Noteworthy Cardiac Surgical Literature in 2013

Introduction

Every year, the understanding of what we do to and for patients in cardiac surgery advances. This year is no exception as several topics were noted that should be understood as they will be commonly referred to or used in the operating room. These topics include the role of coronary bypass in left main and 3-vessel disease and aortic valve repair. In addition to the expanding understanding of literature, there have also been 2 important statements from the Centers for Medicare and Medicaid Services (CMS) regarding the application and approach to both percutaneous valve replacement and left ventricular assist device therapy. As we review the important advances of 2013, we hope this stimulates further discussion and understanding of cardiac surgery in practice today.

Aortic Valve Repair

The introduction of the valve sparing root replacement revolutionized the understanding of the aortic root form and function. The David procedure has become not only an accepted repair technique but also the gold standard for young aneurysm patients with decent valve leaflets. ¹ Familiarity with the David has expanded thought about the potential repair of insufficient aortic valves. de Kerchove and El Khoury have demonstrated that aortic valve repair is more than possible. ² They have created a language to describe the aortic insufficiency based on the Carpentier classification of mitral insufficiency. Furthermore, they have described a systematic approach to repair the specific anatomic deficiencies of the regurgitant aortic valve.

“Principles of Aortic Valve Repair” from March 2013 is one of several publications by this group on their approach to valve repair. If any cardiac surgeon plans to repair aortic valve pathology, they should be familiar. The approach must also be understood by the cardiac anesthesiologist to both describe the anatomic reason for insufficiency and evaluate for failure of repair.

The repair technique is based on the restoration of anatomy at 3 levels. They describe the aortic valve in 2 separate parts. First, the external support of the valve consists of the “functional annulus” made up of the aortic annulus and the sinotubular junction. These are combined because they provide the external framework of the valve. Essentially, the annulus as it would be described by echocardiography and the sinotubular junction need to be of similar diameters to allow for the normal leaflets to coapt sufficiently. Excessive dilatation of either the annulus or the sinotubular junction compromise the symmetry of the cylinder required for adequate valve function. The leaflets make up the second component of the valve. The leaflets need to be compliant and of sufficient height to coapt. They stress that coaptation length can vary based on the annulus size as roots with a larger functional annulus require a greater coaptation reserve, or longer coaptation size, to maintain valvular competence. Leaflet compliance and height will remain important when evaluating these valves.

Using this terminology, the functional classification of aortic insufficiency not only describes the anatomic deficiencies of the valve but can also dictate the method of repair. Analogous to the Carpentier classification for mitral insufficiency, type I disease is due to dilatation of either the sinotubular junction or the annulus. Type 2 disease is caused by cusp prolapse, and type 3 pathology comes from restrictive cusp disease. The principles of repair aim to restore the normal geometry of the functional annulus and leaflets with maintenance of leaflet mobility. The authors do an elegant job of describing repair techniques for these valves ranging from recreating the functional annulus with a David procedure to shortening the leading edge of the prolapsing valve to restore coaptation height. Restricted leaflets can be lengthened with the addition of a patch in the belly of the restricted cusp. The detailed approaches and the specific repair techniques that parallel the anatomy are described in an easily understandable manner, even to simple surgeons. However, the repair of the functional annulus is a much simpler foray than any attempt at leaflet alteration. Despite the complexity of leaflet restoration, the authors do a great job of describing their techniques in simple language.

While aortic valve repair remains in its infancy, some early literature is emerging with regard to 5- to 10-year follow-up from mostly single institution reports. ³ The reports published to date, predominantly from the aforementioned group, show admirable results with 10-year freedom for aortic valve intervention of 89% and 10-year freedom from >2+ AI of 86%. They emphasize that valve-sparing roots probably provide more durable outcomes than lesser procedures that might allow functional annular deterioration over time.

Aortic valve repair is emerging as a durable potential for the avoidance of the issues with the available prosthetic valves. While useful only in aortic insufficiency, limiting the exposure to prosthetic valve anticoagulation or deterioration may be important for these generally young patients. The approach used by Dr El Khoury and colleagues simplifies the decision processes while most likely expanding the cohort of patients eligible for aortic valve repair moving forward.

Coronary Artery Disease: CABG Versus PCI

As percutaneous intervention for coronary artery disease has advanced, the indications for its use over coronary artery bypass has advanced as well. Throughout this evolution, the 3-vessel disease and left main disease have remained surgical indications. The SYNTAX trial was designed to compare percutaneous intervention with first-generation paclitaxel drug-eluting stents to coronary bypass in a randomized fashion. ⁴ Eighty-five centers across the United States and Europe assigned patients 1:1 randomly between the 2 groups. The 1-year and 3-year results have been previously published. The 1-year revascularization rate was increased in the drug-eluting stent (DES) group, while stroke rate was higher in the coronary artery bypass grafting (CABG) group. The 3-year composite of major adverse cardiac and cerebrovascular events (MACCE), myocardial infarction, and repeat revascularization were higher in the DES group. All other endpoints were similar.

The 5-year results were reported in Lancet in early 2013. The study enrolled 1800 patients distributed with 897 CABGs and 903 percutaneous coronary interventions (PCIs). From this split, 805 of the CABGs (90%) and 871 PCIs (97%) completed 5-year follow-up. The demographics of the cohorts were similar with a mean age of 65 years and over 70% being male and equal distributions of diabetes. The SYNTAX scores and average number of vessels intervened on were similar in both groups. As one might predict, the use of dual antiplatelet therapy was significantly higher in the DES cohort than the CABG cohort, but use of aspirin was similar in both groups.

By Kaplan–Meier estimates, MACCE was significantly lower in CABG at 26.9% versus DES at 37.3% (P < .0001). Additionally, the rates of myocardial infarction; the combination of death, stroke, and/or myocardial infarction; and repeat revascularization were all significantly lower in CABG compared with DES. ⁴ The rates of all-cause mortality and stroke were similar between the groups, though. The Kaplan–Meier estimate of cardiac death was significantly worse in DES (5.3% in CABG vs 9% in DES, P = .003).

Subgroup analysis suggested no difference between MACCE in left main disease, but MACCE rates in 3-vessel disease were 50% higher in DES (37.5% compared with CABG 24.2%, P < .0001). In diabetic patients, MACCE was also higher with use of DES. When comparing SYNTAX scores between the groups, outcomes were similar if the score was less than 23, but again CABG was superior to DES with SYNTAX scores >22 with regard to survival, myocardial infarction, and repeat revascularization rates.

The authors, consisting of both surgeons and interventional cardiologists, conclude that CABG remains the standard of care for patients with complex coronary disease. This conclusion is based on improved MACCE, cardiac death, myocardial infarction, and repeat revascularization rates for patients undergoing CABG. This difference is particularly striking for higher complexity PCI as predicted by a SYNTAX score of >22, which comprised two thirds of the patients included in this study. Based on this study, the majority of patients with left main or 3-vessel disease should be treated with CABG, but a role for PCI remains in patients with less complex disease as determined by low SYNTAX scores.

Transcatheter Aortic Valve Therapy

With the FDA approval of the Edwards Sapien Transcatheter Aortic Valve Replacement (TAVR), a revolutionary approach to aortic valve replacement is now possible. Given the call for rational dispersion of this technology, the CMS has issued a strict National Coverage Determination (NCD) for TAVR. ⁵ It will be the purpose of this section to review key features of the NCD for TAVR to highlight this NCD as a likely paradigm for the introduction of new technology for cardiovascular disease.

The CMS issued this NCD in May 2012. ⁵ Key features of this NCD include the following: infrastructure specifications that mandate the presence of an on-site heart surgery program, a cardiac catheterization laboratory or a hybrid operating room/catheterization lab with a fixed radiographic imaging system, and postprocedure intensive care with experienced personnel. Central to this NCD is the concept of the Heart Team to evaluate patients for TAVR. The Heart Team is defined as a cohesive, multidisciplinary team of medical professionals. The heart team typically includes cardiac surgeons, interventional cardiologists, echocardiographers, imaging specialists, and nursing coordinators. Also specifically mentioned in this NCD is that 2 cardiac surgeons must have independently examined the patient face-to-face and deemed them either inoperable or very high risk for conventional open heart AVR. Their evaluation must be documented. The NCD also mandates that the aortic valve implantation system must have received FDA premarket approval. Specific volume criteria based on open heart AVRs and coronary catheterizations are mentioned. The volume criteria differ for programs with prior TAVR experience versus those programs wishing to institute a TAVR program. Finally, it is specified that heart team’s interventional cardiologists and cardiac surgeons must jointly participate in the intraoperative technical aspects of TAVR.

The NCD specifies that all outcomes from TAVR must be reported to a national, audited registry. The registry that fulfills these criteria is the TVT Registry. The TVT registry is conjointly sponsored by the American College of Cardiology (ACC) and the Society of Thoracic Surgeons. The key outcomes tracked in this registry include the following: stroke, all-cause mortality, transient ischemic attacks (TIAs), major vascular events, acute kidney injury, repeat aortic valve procedures, and quality of life. Finally, the NCD will cover TAVR as a non-FDA approved indication when performed within a clinical study. Currently, Partner II, the Core Valve Pivotal Trial, and the Partner III study are examples of such trials.

The TVT registry has already proven to be a dynamic registry capable of tracking and reporting early outcomes. Noteworthy is that reporting to this registry is mandatory; thus, 100% of TAVR cases performed in the United States are captured. A recent publication from this registry reported results of more than 7500 patients who received the commercially approved Sapien transcatheter aortic valve. Of these cases, 20% were deemed “inoperable” for conventional aortic valve replacement and 80% were deemed high risk. The median STS predicted risk of mortality was 7%. The majority of patients underwent transfemoral (TF) access, with 64% undergoing TF access; 29% underwent transapical access (TA); and 7% alternative access—either transaortic or axillary/subclavian. Following TAVR, the in-hospital mortality was 5.5%, with the complications of stroke occurring in 2.0% of patients, dialysis-dependent renal failure occurring in 1.9% of patients, and major vascular injury occurring in 6.4% of patients. ⁶ The results from this early commercial experience of approximately 225 US centers compares very favorably with European and other results with TAVR. Clearly, the TVT registry will be pivotal in tracking the rollout of this very innovative and fast-moving technology.

Mechanical Circulatory Support

The CMS issued an important update to the original VAD NCD from 2003. ⁷ This final decision memo was obtained after the CMS engaged various leaders in heart failure cardiology and cardiothoracic surgery in face-to-face meetings and after a period of public comment. Important changes to the CMS NCD for VADs include the following: regarding VAD implants performed for a bridge to transplant (BTT) indication, the patient must be actively listed on the UNOS heart transplant list (Status 1a, 1b, 2). Status 7 is considered temporarily unsuitable to receive a thoracic organ transplant and does not meet the CMS criteria of “actively” listed. With respect to Destination Therapy patients, the CMS clarified that declaration of patients who are not candidates for heart transplantation is based on information available at the time of VAD implant. This declaration does not preclude patients from becoming transplant candidates with time. Clearly, there is an attempt to recognize that a substantial number of patients implanted with a DT indication, for example, high PVR with pulmonary hypertension, acute renal failure, or social concerns, may have these situations resolved and allow for ultimate transplantation. Noteworthy is the lack of coverage for either BIVAD implantation or RVAD implantation with this latest change in the NCD coverage.

Perhaps most intriguing in the latest coverage memo is that reporting of data to the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) is no longer required for centers to receive reimbursement from the CMS. INTERMACS is the central repository of contemporary registry based data for all FDA-approved VADs and as such has been an integral part of moving the MCS community forward. It remains to be determined whether centers will continue to report to INTERMACS without this explicit requirement in the latest NCD. Other salient features of this latest memo include the mention of a “heart team” concept—emphasizing that heart failure cardiology, CT surgery, a VAD coordinator, social work, and palliative care all are necessary components of a VAD program.

Left Ventricular Assist Device (LVAD) Thrombosis Trend

Over the last couple of years, there has been discussion about potential increase in the rates of device thrombosis following implantation of left ventricular support devices. A recent multicenter publication in the New England Journal of Medicine has delineated the increased thrombosis rate that has gained attention beyond the medical literature and into the lay press. ⁸

The article describes the reference rate of Heartmate II thrombosis as being 2% to 4% based on the pivotal trials and postmarketing approval studies, but single centers began questioning whether this had increased during internal quality reviews. To improve the potential validity of this study, a total of 3 centers were included. This pooled data included 895 Heartmate II LVADs implanted in 837 patients from 2004 to 2013. They explored the primary endpoint of confirmed pump thrombosis. Secondary endpoints included clinical evaluations of suspected pump thrombosis including LDH levels and the outcomes of pump thrombosis management. Median follow-up from implant was 7.8 to 10.7 months, with 278 to 393 patient-years of follow-up.

From this review, 72 pump thromboses were identified in 66 patients. The first month postoperatively was the highest risk of thrombosis (1.4% per month) before dropping off a constant rate of 0.4% per month after the first month. The rate of thrombosis was similar across the 3 hospitals. Estimated occurrences of pump thrombosis were 4.8% at 6 months, 7.5% at 12 months, and 12.3% over 2 years. They found that the rates of pump thrombosis increased drastically around March 2011, with rates of 3 month thrombosis increasing from 2.2% prior to 8.4%. The LDH levels were documented to rise around the same time.

Coupled with the increase in thrombosis, the authors also explored the outcomes after thrombosis. Eleven patients underwent transplantation with a single death 31 days from transplantation. Twenty-one patients underwent pump replacement with 1 death >30 days out, and 1 patient had 3 replacements. Forty thromboses were managed otherwise. Two had pump removal, while 38 were managed with anticoagulation. Half of these died suggesting that medical management is significantly less optimal. Overall, thrombosis more than doubled mortality, compared to no thrombosis, at every time point from 30 to 180 days after thrombosis.

The authors argue that the rate of thrombosis began increasing in 2011 without a plateau to date. INTERMACS data suggested that pump thrombosis rates had increased from 2% prior to May of 2011 to more than 5% after that date. They postulate that INTERMACS data may be limited by voluntary reporting and specific definitions that might apply to this problem. As for the cause of this change, implantation techniques can certainly play a role as do potential relaxation of anticoagulation protocols. However, the change in thrombosis rates seems to be independent of these factors. In the conclusion, the authors feel that better understanding of both patient and pump factors leading to thrombosis is needed. They feel that LVAD therapies remain a required tool for the management of advanced heart failure but that the increased rate of thrombosis must be included in the discussions of patient selection.

Conclusion

The year of 2013 has been an important one for both cardiac surgery research and how these advances can be implemented. Cardiac surgery must continue to adapt and evolve to preserve its place in the treatment of cardiovascular disease moving forward.