Principal Findings:

Of the total of 2,112 patients enrolled in the overall STICH trial, 1,212 were enrolled in hypothesis 1, comparing medical therapy alone (n = 602) to medical therapy + CABG (n = 610), and 1,000 patients in hypothesis 2, comparing CABG + SVR (n = 501) to CABG alone (n = 499).

Hypothesis 1: Baseline characteristics were fairly similar between the two arms. The majority of patients were Caucasian (67%) and about 12% were women. Diabetes was present in about 40% of the patients, prior stroke in 8%, and prior CABG in 3%. About 58% had CCS class I or II angina (37% had none), and about 85% of the patients had NYHA class I or II symptoms.

In the CABG arm, urgent surgery was performed in 5% of the patients, and there was a cross-over of about 17% from the medical therapy arm to CABG; 91% received at least one arterial conduit. Thirty-day mortality was higher in the CABG arm (4% vs. 1%, p = 0.006). Over 6 years of follow-up, the primary outcome of all-cause mortality was similar between the medical therapy + CABG and medical therapy arms (36% vs. 41%, hazard ratio [HR] 0.86, 95% confidence interval [CI] 0.72-1.04, p = 0.12). However, cardiovascular mortality was lower in the medical therapy + CABG arm (28% vs. 33%, HR 0.81, 95% CI 0.66-1.00, p = 0.05), as was all-cause mortality or cardiovascular hospitalization (58% vs. 68%, HR 0.74, 95% CI 0.64-0.85, p < 0.001), and all-cause mortality or repeat revascularization (39% vs. 55%, HR 0.60, 95% CI 0.51-0.71, p < 0.001). No difference for the primary endpoint was noted for any of the subgroups tested, including diabetic patients. On per-protocol analysis, medical therapy + CABG resulted in a significant reduction in all-cause mortality over the duration of follow-up (HR 0.70, 95% CI 0.58-0.84, p < 0.001).

In the subgroup of patients in hypothesis 1 that underwent myocardial viability testing (n = 601), 298 were medically managed, and 303 received medical therapy + CABG. Myocardial viability testing was done with either single-photon emission computed tomography (SPECT) myocardial perfusion imaging or dobutamine echocardiography. Patients with viable myocardium were more likely to survive, irrespective of treatment strategy (33% vs. 50%, p = 0.003) in the unadjusted analysis, but not in the adjusted analysis (p = 0.21). Patients with myocardial viability had a reduction in the composite endpoint of death or or cardiovascular hospitalizations (p = 0.003 in the adjusted analysis). There was no significant interaction between treatment strategy and viability (p = 0.53).

Mitral regurgitation (MR) and outcomes: MR was assessed using a 4-point scale on-site into none or trace (n = 401), mild (n = 493), and moderate/severe (n = 195). In the medical therapy arm, increasing severity of MR was associated with increased long-term mortality at 5 years (30% vs. 47% vs. 55%, p < 0.05 for both). In the subgroup of patients with moderate to severe MR that underwent CABG, 49 patients also underwent mitral valve (MV) surgery (98% repair), and 42 did not. On propensity-matched analysis, patients that underwent concomitant MV surgery had a significant reduction in long-term mortality, as compared with those who did not (41% vs. 55%, HR 0.45, p < 0.05).

10-year results (STICH-Extended study [STICHES]): All-cause mortality for CABG + medical therapy vs. medical therapy alone: 58.9% vs. 66.1%, HR = 0.84, 95% CI 0.73-0.97, p = 0.02; NNT = 14). CV mortality: 40.5% vs. 49.3%, HR = 0.79, 95% CI 0.66-0.93; p = 0.006); all-cause mortality or MI: 61.6% vs. 67.9%, p = 0.03. Effect modification for primary endpoint observed based on angiographic severity of disease: patients with three-vessel disease had an even greater benefit (p for interaction = 0.04). On analyzing data based on age quartiles (<54, >54-≤60, >60-≤67, ≥67 years), older patients had more comorbidities such as type 2 diabetes mellitus and chronic kidney disease. All-cause mortality tended to be higher in older compared with younger patients in both arms, while CV mortality was not statistically significantly different across the spectrum of age. The benefit of CABG + medical therapy vs. medical therapy on all-cause mortality decreased with increasing age (47.5% vs. 60%; 53.1% vs. 59.2%; 66.7% vs. 68.1%; 68.1% vs. 78.6%, p _interaction = 0.062), whereas the benefit of CABG on CV mortality had a U-shaped distribution (35.0% vs. 49.4%; 42.7% vs. 47.4%; 46.5% vs. 47.4%; 38.7% vs. 53.1%; p _interaction = 0.307).

Hypothesis 2: No significant differences in baseline clinical characteristics were present between groups. However, more arterial conduits were utilized in patients undergoing CABG alone. SVR added a median of 27 minutes of cardiopulmonary bypass time to the procedure. A greater reduction in LV end-systolic volume index was observed in the SVR group (-19% vs. -6%, p < 0.001). However, there was no significant difference in the primary endpoint of death and cardiac hospitalization (58% vs. 59%, p = 0.90). There was no greater improvement in NYHA heart failure classification or CCS angina classification with SVR in addition to CABG.

Quality of life, as assessed by the Kansas City Cardiomyopathy Questionnaire, Seattle Angina Questionnaire, and Center for Epidemiological Studies Depression Scale, was not significantly different between groups. In the US cohort, medical costs were greater with CABG + SVR than with CABG alone ($70,717 vs. $56,122, p = 0.004).

On subgroup analysis, patients with better functional status prior to randomization, as evidenced by an ability to walk >300 m (HR 0.77, 95% CI 0.59-0.99, p = 0.038) or a physical activity score >55 (HR 0.79, 95% CI 0.62-1.01, p = 0.06), appeared to derive a benefit from CABG over medical therapy, although tests for heterogeneity were nonsignificant (p = 0.21 and p = 0.41, respectively).

Patients with evidence of angina did not have higher mortality with CABG or medical therapy. Patients undergoing CABG reported greater improvement in anginal symptoms compared with medical therapy alone (OR 0.70, 95% CI 0.55-0.90, p < 0.01).

Sudden cardiac death (SCD) in this patient population: Among patients without an implantable cardioverter-defibrillator (ICD; n = 1,411), 8% died from SCD and 22% from other causes. The 1-, 3-, and 5-year cumulative incidence of SCD after CABG was 2.8%, 6.1%, and 8.5%, respectively. The conditional risk of SCD per month was different: 0.35% in the first 30 days after CABG, and 0.43% between 31 and 90 days. After 6 months, the risk per month decreased to 0.14% and remained stable after that. Variables that were predictive of SCD on multivariable modeling were end-systolic volume index, B-type natriuretic peptide, Duke CAD score, history of atrial fibrillation/flutter (all with increased risk), statin use, serum sodium, and received CABG + SVR (all with lower risk).