The Rationale and Design of the Surgical Treatment for IsChemic Heart failure (STICH) Trial

Abstract

Objectives

The rationale and design of the Surgical Treatment for Ischemic Heart Failure (STICH) trial is described. Prior to STICH, <1000 ischemic cardiomyopathy patients had been studied in randomized comparisons of medical therapy (MED) versus coronary artery bypass graft surgery (CABG). Trial data reflect how these therapies were delivered over 20 years ago and do not indicate the relative benefits of MED versus CABG in contemporary practice.

Methods

Randomization of consenting patients with heart failure, left ventricular (LV) ejection fraction ≤0.35, and coronary artery disease is based on whether patients are judged by attending physicians to be candidates only for CABG or for MED or CABG. Patients eligible for surgical ventricular reconstruction (SVR) due to significant anterior wall akinesis or dyskinesis, but ineligible for MED are randomly assigned to CABG with or without SVR. Patients eligible for MED are randomly assigned between MED only and MED with CABG. Patients eligible for all 3 are randomly assigned evenly to MED only, MED and CABG, or MED and CABG and SVR. Major substudies will examine quality of life, cost-effectiveness, changes in LV volumes, impact of myocardial viability, selected biomarkers, and selected polymorphisms on treatment differences

Conclusions

STICH is an NHLBI-funded multicenter international randomized trial addressing 2 specific primary hypotheses: 1) CABG with intensive MED improves long-term survival compared with MED alone; and 2) in patients with anterior LV dysfunction, SVR to a more normal LV size plus CABG improves survival free of subsequent hospitalization for cardiac cause when compared with CABG alone.

The revascularization of coronary artery disease (CAD) patients with left ventricular (LV) systolic dysfunction (SD) and symptomatic heart failure (HF) has been the subject of much debate and surprisingly little research over the past 30 years.^1,2 Our current understanding of how best to treat these patients stems from subset analyses of the coronary artery bypass grafting (CABG)-medicine clinical trials performed in the 1970s and early 1980s, and analyses of large registries of patients from the same era.^3-7 While such analyses have generally demonstrated that patients with more advanced CAD and more severe LV dysfunction derive larger benefit from CABG relative to medical therapy (MED), in practice, both cardiologists and surgeons have substantial uncertainty about whether these projected benefits are counterbalanced by the increased early risks of the surgical approach.

In 2002, the National Heart, Lung, and Blood Institute (NHLBI) funded the Surgical Treatment for IsChemic Heart failure (STICH) trial to address 2 pressing clinical and policy questions regarding the management of HF patients with surgically revascularizable CAD and depressed LV function: 1) is contemporary CABG surgery superior to contemporary medical/secondary prevention therapy in prolonging survival in these patients; and 2) among patients with significant anterior wall dysfunction, does the addition of surgical ventricular reconstruction (SVR) to CABG improve hospitalization-free survival?

Study Design

Philosophy of Study Design

In the absence of definitive data on the value of CABG in high-risk patients with LVSD, wide diversity exists among providers about how to select patients. Moreover, many patients who might be CABG candidates also have dominant anterior akinesia or dyskinesia that might be reasonable to reconstruct surgically at the time of CABG.⁸ Therefore, the STICH study was designed to let physicians first determine whether potential STICH patients were amenable to CABG after their routine clinical assessment. A threshold LV ejection fraction (EF) of ≤0.35 was established with no lower limit set to preclude study entry. Physicians are encouraged to use any and all cardiac testing necessary to decide whether an individual patient is a candidate for CABG and/or SVR. This philosophy encourages the use of standard practice to identify patients for whom responsible physicians are at equipoise about MED, CABG, and CABG with SVR, and ensures that the STICH cohort has characteristics that now pose the greatest uncertainty in clinical decision-making for patients with ischemic cardiomyopathy. Moreover, all information now commonly used by clinicians in deciding on surgical treatment for a subset of ischemic cardiomyopathy patients will be available in order that the value added by each component to the decision-making process can be evaluated (Table 1).

Table 1.

Major STICH hypotheses

Primary hypotheses
H1/Coronary revascularization hypothesis  Improvement in myocardial perfusion by CABG combined with MED improves longterm  survival compared with MED alone
H2/Left ventricular restoration hypothesis  In patients with dominant anterior wall LV akinesia or dyskinesia, LV shape and size  optimization by SVR combined with CABG and MED improves long-term survival  free of cardiac hospitalization compared with CABG and MED without SVR
Major secondary hypotheses
Presence and extent of dysfunctional but viable myocardium as defined by radionuclide imaging and/or dobutamine stress echo will identify patients with greatest survival advantage of MED and CABG compared with MED alone Echo assessment of LV systolic and diastolic dysfunction, cardiac hemodynamics, and valvular regurgitation at baseline will define patient subgroups that derive substantial improvement or incur substantial risk in allcause mortality if treated with MED and CABG compared with MED alone Surgical intervention with CABG will lead to greater sustained improvements at 4 and 24 months in echo derived indices of LV systolic and diastolic dysfunction, hemodynamics, and valvular regurgitation when compared with MED alone Assessment of myocardial ischemia at baseline by radionuclide perfusion imaging during rest and stress will identify patients with a survival benefit of MED and CABG compared with those treated with MED alone MED and CABG with SVR will lead to more improvement in LV end systolic volume index measured serially at baseline, 4 months, and 24 months compared with MED and CABG without SVR CMR assessment of baseline LV sphericity index will add independent prognostic information beyond LV end systolic index regardless of treatment assignment, and MED and CABG with SVR will lead to improved LV sphericity index when compared with MED and CABG alone Extent of neurohormonal and pro-inflammatory cytokine activation will identify the patient subgroup most likely to derive substantial survival benefit from MED and CABG with or without SVR compared with MED alone Presence of a favorable genotype (e.g., low ACE gene expression/high adenosine expression) will identify the patient subgroup most likely to have long-term improvement in cardiovascular morbidity and mortality MED and CABG will lead to incremental cost effectiveness assessed as cost/life year added and cost/quality-adjusted life year added relative to MED alone MED and CABG will lead to greater improvement in health-related quality of life

H1 = hypothesis 1; CABG = coronary artery bypass grafting; MED = medical therapy; H2 = hypothesis 2; LV = left ventricular; SVR = surgical ventricular reconstruction; ACE = angiotensin-converting enzyme.

Subjects meeting the broad inclusion criteria of CAD amenable to CABG with an LVEF of ≤0.35, without a specific exclusion, are segregated into 3 strata (A, B, and C) depending on investigator-determined suitability for continued MED alone and eligibility for SVR (Fig. 1). Eligibility for MED alone is defined by the investigator, but generally excludes patients with an intraluminal left main coronary artery stenosis of ≥50% or severe, disabling angina (Canadian Cardiovascular Society Class ≥III) unresponsive to non-surgical interventions. Eligibility for SVR is defined as dominant LV akinesia or dyskinesia amenable to SVR. Stratum A subjects are defined as suitable for MED with or without CABG, and consenting patients are randomly assigned in a 1:1 ratio between MED alone or with CABG. Stratum B subjects, defined as eligible for all 3 treatment options, are randomly assigned 1:1:1 to either MED alone, MED with CABG, or MED with CABG and SVR. Subjects eligible for CABG with and without SVR are randomly assigned 1:1 in Stratum C to either CABG or CABG with SVR. After stratum eligibility is established and informed consent is obtained, treatment allocation is made to a specific therapy based on an undisclosed permuted block randomization scheme.

Figure 1.

STICH stratum and treatment assignment.

Study Population

The STICH trial is designed to enroll at least 2000 men and women aged ≥18 years who have CAD amenable to revascularization and LVSD defined by a clinically-determined LVEF of ≤0.35. STICH trial entry criteria are summarized in Table 2. Patients awaiting a planned PCI to treat symptomatic CAD within the next 30 days are not eligible, although previous PCI is not an exclusion. While planned operative treatment of the aortic valve excludes potential candidates, the decision to pursue operative management of any other valves, specifically the mitral valve, is left to the discretion of responsible physicians and surgeons.

Table 2.

STICH trial entry criteria

Inclusion criteria*
Men Women not of childbearing potential Age ≥18 years LVEF ≤0.35 measured by contrast magnetic resonance ventriculogram, gated SPECT ventriculogram, echo, or contrast ventriculogram within 3 months of trial entry CAD suitable for revascularization
Exclusion criteria †
Failure to provide informed consent Aortic valvular heart disease indicating need for aortic valve repair or replacement Cardiogenic shock (within 72 hrs of randomization); defined by need for IABP support or requirement of IV inotropic support Plan for PCI of CAD Recent acute MI judged to be an important cause of LV dysfunction History of more than 1 prior CABG Non-cardiac illness with a life expectancy of <3 yrs Non-cardiac illness imposing substantial operative mortality Conditions/circumstances likely to lead to poor treatment adherence (e.g., history of poor compliance, alcohol or drug dependency, psychiatric illness, no fixed abode) Prior heart, kidney, liver, or lung transplant Current participation in another clinical trial in which patient is taking an investigational drug or receiving an investigational medical device
MED therapy eligibility criteria
Absence of left main CAD defined by intraluminal stenosis ≥50% Absence of Canadian Class III angina or great (angina markedly limiting ordinary activity)
SVR eligibility criterion
Dominant akinesia or dyskinesia of anterior LV wall amenable to SVR

^*Patients may qualify for inclusion in the study.

^†None of these may exist at randomization.

LV = left ventricular; EF = ejection fraction; SPECT = single photon emission computed tomography; CAD = coronary artery disease; IV = intravenous; IABP = intraaortic balloon pump; PCI = percutaneous coronary intervention; MI = myocardial infarction; CABG = coronary artery bypass grafting; MED = medical therapy; SVR = surgical ventricular reconstruction.

Baseline and Follow-up Studies

Following informed consent, baseline demographics, physical examination, laboratory data, and medical including procedural history and details are collected and all patients undergo any remaining requisite baseline studies (Table E1). Patients eligible for MED alone enrolled into Strata A and B undergo baseline radionuclide perfusion and viability imaging and a modified Bruce protocol exercise stress test, if they are able to exercise. SVR eligible patients enrolled into Strata B and C preferably undergo baseline cardiovascular magnetic resonance (CMR) or gated single photon emission computed tomography (SPECT) perfusion imaging with follow-up at 4 months and 24 months to assess post-operative size, shape, and function.

All STICH subjects will undergo echocardiography (echo) and blood sampling for neurohormonal, cytokine, and genetic analyses, a detailed quality of life assessment, and a 6-minute walk test, if appropriate based on subject status. These baseline studies are repeated in 4 months with a further repeat echo performed at 24 months. Continuing assessment of quality of life and exercise capacity for all patients will occur at yearly intervals post-randomization. Table E2 outlines STICH follow-up studies. Details of core laboratory testing protocols are available at www.stichtrial.org.

Medical Therapy

The STICH Medical Therapy Committee is mandated to regularly review evidence and redefine optimal MED when necessary for all STICH subjects regardless of randomization strata or treatment. Unless contraindicated, optimal MED includes angiotensin-converting enzyme inhibitor and/or angiotensin receptor blocker, beta-blocker, aldosterone antagonist, and antiplatelet agents adjusted to optimal doses within 30 days post-randomization. HMG-CoA reductase inhibitor, diuretic, and digitalis use will be individualized to patient-specific indications. The use of implantable defibrillators is encouraged as part of MED and should be used in compliance with standard guidelines. A lead cardiologist at each site is responsible for ensuring high-quality MED for all patients including those treated with surgery.

Surgical Therapy

Subjects randomly assigned to either CABG or CABG with SVR will receive the protocol-determined intervention no later than 14 days after randomization. CABG is performed using at least one internal mammary conduit unless unavailable or inadequate. Use of cardiopulmonary bypass for CABG is left to the discretion of the surgeon. Patients with secondary mitral regurgitation judged to require mitral valve repair or replacement may undergo this procedure, although this is not mandated by protocol. The SVR typically occurs after CABG by any acceptable reconstruction method that consistently increases LVEF and decreases end systolic volume. The general operative technique for SVR has been previously described.⁸ Use of a sizing device to judge appropriate LV chamber size and the decision of whether and how to patch the LV closure site are left to the operating surgeon.

Before and during hypothesis 2 (H2) patient recruitment, multiple educational opportunities will be made available to cardiac surgeons to refine surgical decision-making and operative techniques for SVR patients. A lead cardiac surgeon at each site is responsible for initially certifying that all STICH surgeons meet qualification standards set by the Surgical Therapy Committee, and for maintaining a high quality of surgical care for the duration of the trial. The most experienced cardiac surgeons at each site are certified as eligible to operate on randomized patients. The minimum requirement for certification is evidence of 25 CABG patients with LVEF ≤0.40 who were operated on with ≤5% mortality. Before cardiac surgeons are certified to perform SVR on a randomized patient, they are required to perform at least 5 SVR procedures without a perioperative death and demonstrate consistent LV volume reduction after operation.

Patient Follow-up

The first follow-up for clinical status occurs at hospital discharge or at 30 days after randomization, whichever comes first. Participants are subsequently seen at 4-month intervals post-randomization for the first year, and no less than every 6 months after year one. At these visits, interval assessments of HF and angina symptom status, current use of medications, and clinical endpoint data including hospitalizations and procedures since the previous visit are documented. Regardless of therapy received, all study participants are followed in this manner until study completion.

Statistical Issues

Hypothesis 1 Power

All patients enrolled into Stratum A and two-thirds of patients enrolled into Stratum B, those assigned to MED with or without CABG, comprise the hypothesis 1 (H1) study cohort. All-cause mortality is the primary endpoint of H1. H1 power estimates were based on an estimated 3-year mortality rate of 25% in the MED arm of the trial. This rate is slightly lower than the mortality observed in the control arm (no implantable cardioverter defibrillator [ICD] implanted) of the ischemic cardiomyopathy cohort of the SCD-HeFT Trial.⁹ Approximately 400 deaths are required to achieve 90% statistical power for detecting a treatment difference consisting of a 25% reduction in mortality with MED with CABG. These 400 deaths are projected to occur if at least 1000 H1 patients are followed for an average of 6.5 years or 1200 patients are followed for an average of 5.5 years.

Hypothesis 2 Power

All patients randomly assigned to Stratum C and the two-thirds of patients entered into Stratum B assigned to either CABG or CABG with SVR comprise the H2 study cohort. A composite endpoint of survival free of cardiac hospitalization was chosen for H2 since no data exist to suggest that adding SVR to CABG improves survival over CABG alone. Moreover, this composite endpoint has validity for patients who would be likely to consent to adding SVR to a planned CABG. The planned enrollment of 1000 patients into H2 provides a 90% power to detect a 20% reduction in mortality and cardiac hospitalization by the addition of SVR to CABG, assuming that the 3-year event rate for those treated with CABG alone is 45% or higher.

Treatment Crossovers

Crossovers between MED to CABG or CABG and SVR are anticipated within the first year due to the dynamic nature of coronary disease, and the potential for deteriorating symptoms unresponsive to standard MED and requiring symptomatic relief. Patients randomly assigned to CABG and SVR may also not receive SVR due to intraoperative decisions that maximize patient safety. PCI is not regarded as a treatment crossover but rather as downstream medical care associated with any of the treatment strategies tested in STICH. The sample sizes for both H1 and H2 allow for treatment crossovers of as much as 20% without sacrificing power so long as the control event rates remain as projected.

Secondary Endpoints

Secondary endpoints for both hypotheses include all-cause mortality at 30 days, cardiovascular mortality, survival free of HF hospitalization, survival free of subsequent revascularization, need for cardiac transplantation, need for LV assist device, all-cause hospitalization, LV size, LV function, total health care costs, cost-effectiveness, and quality of life. Other important efficacy parameters that will be evaluated include fatal and nonfatal MI, fatal and non-fatal stroke, sudden death with or without resuscitation, survival free of PCI, and survival free of CABG.

Approach to Data Analysis

The primary efficacy analyses will be performed according to the principle of intention to treat; that is, patients will be analyzed (and endpoints attributed) according to the treatment arm to which they were randomly assigned regardless of subsequent crossover or non-adherence to the assigned treatment. Statistical comparisons will be performed using 2-sided significance tests. For the primary comparisons of MED versus CABG in H1 and CABG versus CABG with SVR in H2, the log-rank test will be used adjusting for the stratum in which the patients were enrolled.¹⁰ Cumulative event rates will be calculated according to the Kaplan-Meier method.¹¹ Event (or censoring) times for all patients will be measured from the time of randomization (time zero). Relative risks will be expressed as hazard ratios with associated confidence intervals and will be derived from the Cox proportional hazards model.^12,13 The Cox model will also be used in the assessment of treatment differences and analyses for many of the secondary endpoints. Interim analyses of the data will be performed and reviewed by an independent Data and Safety Monitoring Board (DSMB) appointed by the NHLBI. Interim treatment comparisons will be monitored with the use of 2-sided, symmetric O’Brien-Fleming boundaries generated with the Lan-DeMets alpha-spending function approach to group-sequential testing.^14,15

Study Leadership

The STICH trial is an investigator-initiated, international study funded by the NHLBI of the National Institutes of Health. The STICH Steering Committee is comprised of principal investigators at all enrolling sites. An executive council is empowered by the Steering Committee to make day-to-day decisions (Appendix E1). However, all major executive council decisions are subject to review and approval of the Steering Committee.

Trial operations, site management and monitoring, statistical planning, and data analysis are being coordinated at the Duke Clinical Research Institute of Duke University in Durham, North Carolina. An endpoint classification committee blindly adjudicates all hospitalizations during follow-up. This committee functions independently of the operational team and is chaired by a cardiovascular specialist without direct access to potential STICH patients. A fully independent DSMB (Appendix E2) has been empowered to review unblinded safety and efficacy data no less than twice yearly using the pre-specified early termination rules described in the previous section.

Discussion

Rationale for Contemporary Evaluation of Survival Benefit Associated with Addition of CABG to MED in Patients with Ischemic Cardiomyopathy

Randomized trials of CABG surgery enrolling 2234 patients between 1971-1979 established the safety of surgical revascularization in patients with preserved LV systolic function and chronic stable angina.⁵ Improved survival after CABG compared with MED alone was shown for patients with angina and flow-limiting stenoses of the left main coronary artery or multiple coronary arteries, especially in patients with severe stenosis of the proximal left anterior descending coronary artery. The Coronary Artery Surgery Study (CASS) was the only one of these 7 early studies to stratify randomization of the 780 patients enrolled based on LVEF (0.35-0.50 vs. >0.50), angiographic extent of CAD (1-, 2-, 3-vessel stenosis), and presence or absence of angina at the time of enrollment. At the conclusion of 10 years of follow-up, 82 (21%) of 390 patients randomly assigned to receive MED alone had died and 70 (18%) of 390 patients randomly assigned to receive MED plus CABG had died (P = .25).¹⁶ The subgroup of patients with an LVEF of 0.35-0.50 was the only stratified subgroup in CASS to show a significant survival difference.³ Of the 160 patients in CASS who had LVSD, 32 (38%) of the 82 MED-treated patients died while 16 (21%) of the 78 patients who underwent CABG (P = .01) died. However, the 54-patient subset of CASS who had LVSD but no angina did not derive a statistical survival advantage from CABG (P = .12).

Since the initial description of the clinical syndrome of ischemic cardiomyopathy over 3 decades ago,¹⁷ the clinical care of patients with CAD and LVSD has undergone a dramatic evolution. When the last patient was randomized between CABG and MED during the time of CASS enrollment, MED for patients with HF, LVSD, and CAD was limited to digitalis and diuretics, which are medications now known to have a neutral impact on mortality. Current American College of Cardiology/American Heart Association guidelines highlight the major advances in pharmacotherapy and device therapy that have improved the quality of life and survival of patients with CAD, HF, and LVSD.¹⁸ However, the evidence base remains deficient in identifying which, if any, patients with CAD and LVSD should receive revascularization. While specific clinical problems in this population, such as severe angina, are used to decide on revascularization strategies, the vast majority of patients with ischemic cardiomyopathy have limited or no angina and fall into a gray zone where clear evidence for adding CABG to MED is either absent or outdated. Thus, divergent views have evolved among clinicians since the reports of the CASS data as to the most appropriate diagnostic and management strategy for patients with ischemic cardiomyopathy. Table E3 summarizes current guideline recommendations as they relate to the selection of CABG as a treatment option for patients with poor LV function.^18-22

The weight of observational data suggests that HF and LVSD remain associated with a higher risk of post-CABG complications and mortality, and while increasing numbers of patients with LVSD and angina are referred to CABG, it may remain underutilized.^23,24 Of the 24,959 patients screened and entered into the CASS registry from 1975-1979, 751 patients had severe LVSD as defined by an LVEF of <35%, 231 patients underwent CABG, and 420 patients remained on MED.⁴ Overall 5-year survival favored MED. However, the subset of patients with an LVEF of <26% had a survival advantage demonstrated for the 82 patients who received CABG compared with the 172 patients treated medically (P = .006). The CASS registry CABG group had more angina and less severe LVSD than the CABG cohort in the randomized trial. The CASS registry CABG patients presenting with angina had improved survival free of functional limitation, but patients with predominant HF did not. A large observational series summarized outcomes for 1391 patients with CAD and an LVEF of <40% who underwent diagnostic coronary angiography from 1969-1994.⁷ After adjustment for baseline differences, the 339 patients who received CABG had a better survival rate than the 1052 patients who received only MED. However, unlike CASS registry patients, this observed survival advantage occurred in all patients regardless of LVEF and HF status. The perspective that CABG use in patients with LVSD should be confined only to the patient subset with ongoing angina continues to influence treatment guidelines today. Should STICH show that revascularization provides survival benefit beyond that of modern MED, not only will CABG be used more aggressively in patients with LVSD and CAD amenable to CABG, but it will indicate the need to evaluate for CAD in all patients who present with HF and LVSD.

Rationale for a Prospective Assessment of Myocardial Viability

Viability testing by various modalities is currently used by many clinicians to select patients for surgery. Positron emission tomography has been considered by many to be the gold standard, but high cost and lack of widespread availability has led to limited use.²⁵ Thallium or technetium based nuclear scintigraphy, using one of many rest or stress imaging protocols, has become more widespread and has acceptable sensitivity and specificity.²⁶ Because it is readily available, dobutamine stress echocardiography to demonstrate augmentation of contractile response has become the test of choice for assessing tissue viability at many institutions.²⁷ Although not as widely available, delayed contrast enhancement on CMR scan is a reliable method to detect non-viable myocardium.²⁸ While guidelines recommend that the presence or absence of viable myocardium may be considered in the decision as to whether revascularization should be recommended,^18-21 the strength of this recommendation varies among the individual guidelines (Table E3). Therefore, extensive heterogeneity exists on how clinically available structural and functional imaging is incorporated into treatment decisions for patients with CAD and LVSD.

Viable myocardium in patients with LVSD and CAD appears to predict contractile recovery in dysfunctional myocardial segments regardless of whether patients receive MED alone or CABG.²⁹ While no studies have validated the use of viability testing regardless of the imaging modality in a prospective randomized trial with a survival endpoint, many retrospective non-randomized observational series have been published that evaluated the association between results of viability testing and short-term clinical outcomes. Systematic reviews^30,31 suggest that, when present, myocardial viability is associated with improved short-term (18 month) survival in patients referred to CABG, but conversely the absence of viability is not necessarily associated with disparate outcomes between medically and surgically treated patients. Unfortunately, a multitude of confounding factors including the lack of standard definitions for viability, the heterogeneous non-randomized populations studied, and the lack of uniform MED received, limit the applicability of this data to current practice. While the use of viability imaging is predicated on an ability to predict functional recovery, failure to improve LV function is not necessarily associated with a worse clinical outcome.³² Since CABG in patients with LVSD continues to be associated with substantial operative risk, when and how viability imaging is used to reliably predict long-term outcomes are critical questions that require definitive answers.

Rationale for SVR as an Adjunct to CABG for Patients with Predominant Anterior Akinesia

Surgical reconstruction of akinetic, dyskinetic, or aneurysmal segments may theoretically lower LV wall stress, myocardial oxygen consumption, and stroke volume while preserving or improving contractile function in the remaining ventricle. In the CASS registry population of LVSD, over 30% of the patients who underwent CABG received concomitant ventricular reconstruction surgery.⁴ These early linear plications or resections of dyskinetic scar commonly deformed the LV cavity into a box-like shape and did not consistently improve ventricular performance.³³ Intracavity reconstruction techniques were developed for repairing defects left by aneurysm resection that reduced LV cavity size but retained a more elliptical configuration of the ventricle.³⁴ Dor and colleagues⁸ advocated the use of SVR, not only in patients with dyskinetic scar, but also in those with only akinetic myocardial segments. A preserved epicardial covering of myocardial fibrosis may make these akinetic zones appear normal at the time of cardiac operation, but palpable thinning usually can be appreciated in the arrested decompressed heart. Unlike earlier LV aneurysmectomy that removed myocardial scar or the Batista operation³⁵ that reduced LV size by excision of portions of the LV wall, the objective of the SVR operation is to reshape and decrease the size of the LV by decreasing the circumference of the endocardial scar through an incision in normal epicardium. Myocardial scar is occasionally excised during SVR, but commonly no tissue is removed at operation. The intrinsic scar or an extrinsic patch may be used during closure of the left ventriculotomy to decrease linear wall tension and avoid the restrictive physiology of under-sizing the left ventricle.

The RESTORE registry group reported on 1198 patients who underwent SVR at 11 centers with a 5.1% operative mortality and an 88% 18-month survival rate for all patients.³⁶ A recent report from the Society of Thoracic Surgery database³⁷ suggests that SVR is being performed with increasing frequency for the treatment of patients with HF, CAD, and LVSD; however, the perioperative risk may be more substantial. In this 2002-2004 U.S. sample, 731 patients underwent SVR at 141 out of 576 reporting centers. The perioperative 30-day outcomes were 9.3% for mortality and 33.5% for any major complication. Use of SVR as an adjunct to CABG for patients with LVSD cannot yet be justified on firm evidence since no reports are available comparing outcomes of CABG and SVR with CABG alone in similar populations. H2 of the STICH trial was designed to address this important question.

Current Status of the STICH Trial

The first patient was randomized into STICH in July 2002. Since then, the STICH steering committee has approved 3 protocol amendments to facilitate successful completion of trial enrollment (Appendix E3). Randomization of 1000 planned subjects into H2 was completed in January 2006 and led to the closure of Strata B and C to further enrollment. This represents the largest ever randomized comparison of 2 cardiac surgical strategies. H2 primary endpoint results are expected to be published in 2009. In June 2006, STICH enrollment into H1 surpassed the CASS study as the largest ever comparison of cardiac surgical and medical approaches to chronic CAD. H1 is expected to conclude enrollment of at least 1200 patients by early 2007. Published H1 results are anticipated by 2011.