# Stents vs. Bypass Surgery for Three-Vessel Heart Disease: What the FAME 3 Study Found Patients with severe blockages in all three major heart arteries face an important treatment decision: minimally invasive stenting or open-heart bypass surgery. The FAME 3 study followed 1,500 such patients for one year and found that 10.6% of those who received stents guided by fractional flow reserve (FFR) experienced a major heart-related event, compared with 6.9% of those who underwent bypass surgery. This difference meant that the stent procedure did not meet the study's pre-defined criteria to be considered "not worse than" surgery. However, patients who received stents had significantly fewer short-term complications, including major bleeding, irregular heart rhythms, and kidney injury. # Stents vs. Bypass Surgery for Three-Vessel Heart Disease: What the FAME 3 Study Found ## Table of Contents - Key Points - Background: Why This Study Matters - Understanding the Two Treatment Approaches - How the Study Was Designed (Methods) - Who Took Part in the Study - What Treatments Patients Actually Received - Key Findings: The Primary Result - Detailed Results: Individual Events - Safety: Short-Term Complications - Subgroup Analyses: Who Benefits From Which? - What This Means for Patients - Study Limitations - Recommendations for Patients - Frequently Asked Questions - Source Information ## Key Points - In 1,500 patients with three-vessel disease, bypass surgery led to fewer major heart events at 1 year than FFR-guided stenting: 6.9% versus 10.6%. - Stenting failed to meet the noninferiority threshold; the hazard ratio was 1.5 (95% CI 1.1–2.2), favoring surgery. - Bypass surgery had more short-term complications: major bleeding 3.8% vs 1.6%, arrhythmia 14.1% vs 2.4%, and higher 30-day readmissions. - FFR-guided stenting avoided stenting 24% of blockages that did not limit blood flow, yet still resulted in more repeat revascularizations. - Results were consistent across subgroups including age, sex, diabetes, and SYNTAX score; no subgroup favored stenting over surgery. ## Background: Why This Study Matters Coronary artery disease occurs when plaque builds up inside the arteries that supply blood to the heart muscle. When blockages develop in all three of the major heart arteries — a condition called **three-vessel coronary artery disease** — the heart's blood supply is seriously compromised. Without treatment, patients face a high risk of heart attacks, heart failure, and death. For decades, doctors have known that **coronary-artery bypass grafting (CABG)**, commonly called bypass surgery, produces better long-term outcomes than **percutaneous coronary intervention (PCI)**, commonly called angioplasty or stenting, for patients with three-vessel disease. Earlier large, randomized trials consistently showed that patients who received bypass surgery had lower rates of death, heart attacks, and the need for repeat procedures compared with those who received stents. But those older trials had two important limitations. First, they rarely used second-generation drug-eluting stents — newer stents that slowly release medication to prevent scar tissue from blocking the artery again. Second, they did not routinely use a measurement called **fractional flow reserve (FFR)** to guide stent placement. The FAME 3 trial was designed specifically to address these gaps and to answer a modern question: With today's best stent technology and FFR guidance, can stenting match the results of bypass surgery? Previous research had already shown that second-generation drug-eluting stents lead to lower rates of stent thrombosis (blood clots inside the stent), procedure-related and spontaneous heart attacks, re-narrowing of the treated artery (restenosis), and death compared with first-generation stents. Similarly, FFR-guided PCI produces better short-term and long-term outcomes than angiography-guided PCI (where decisions are made based only on X-ray images) or medical therapy alone. ## Understanding the Two Treatment Approaches To understand this study, it helps to know how the two procedures differ. **Coronary-artery bypass grafting (CABG)** is open-heart surgery. The surgeon takes a healthy blood vessel from another part of the body — most commonly the left internal thoracic artery from the chest wall, or a leg vein — and attaches it to the heart artery beyond the blockage, creating a "bypass" so blood can flow around the narrowed area. In this study, 97% of patients who had CABG received a left internal thoracic artery graft, which is considered the gold-standard bypass vessel. **Percutaneous coronary intervention (PCI)** is far less invasive. A thin tube called a catheter is threaded through an artery in the wrist or groin up to the heart. A tiny balloon is inflated inside the narrowed artery to compress the plaque, and a mesh tube called a **stent** is left in place to keep the artery open. The procedure requires no surgical incision in the chest and typically involves a hospital stay of only a few days. **Fractional flow reserve (FFR)** is a technique used during PCI. A very thin wire with a pressure sensor is passed into the heart artery. The wire measures blood pressure just before and just after a blockage while the heart is stressed with a medication called adenosine. The ratio of these pressures — the FFR value — tells doctors how severely the blockage restricts blood flow. An FFR of 1.0 means no obstruction at all, while lower numbers mean more significant blockage. In this study, and in standard clinical practice, a cutoff of **0.80 or lower** was used to decide that a blockage was significant enough to deserve a stent. This approach prevents unnecessary stenting of blockages that are not actually limiting blood flow — blockages that respond just as well to medications alone. ## How the Study Was Designed (Methods) The FAME 3 trial (formally called the **Fractional Flow Reserve versus Angiography for Multivessel Evaluation 3** trial) was an investigator-initiated, multicenter, international, randomized, controlled trial. The study was conducted at 48 medical centers across multiple countries. Patients with three-vessel coronary artery disease that did not involve the left main coronary artery were randomly assigned in a 1:1 ratio to receive either CABG or FFR-guided PCI — much like flipping a coin to decide which treatment each patient received. Randomization was performed using a Web-based system and was **stratified according to diabetes status and trial site**, which means researchers made sure the two groups were balanced in terms of these important factors. The major inclusion criterion was the presence of three-vessel coronary artery disease, defined as **at least 50% narrowing (diameter stenosis)** of each of the three major epicardial arteries or their major side branches, as judged by visual inspection of the angiogram. The narrowing also had to be treatable by either PCI or CABG, as determined by a "heart team" at each trial site. Patients were excluded if they had a recent **ST-segment elevation myocardial infarction (STEMI)** — the most serious type of heart attack — or **cardiogenic shock** (when the heart fails to pump enough blood to the body). Patients with a **left ventricular ejection fraction of less than 30%** (severely weakened heart-pumping function) were also excluded. For patients assigned to PCI, the protocol required that FFR be measured using a coronary pressure wire (Abbott Vascular) with intravenous or intracoronary adenosine. Only blockages with an FFR of **0.80 or lower** were to be treated with stents. The stents used were current-generation **zotarolimus-eluting stents** (Resolute Integrity or Resolute Onyx, made by Medtronic). These stents slowly release the drug zotarolimus to prevent the artery from re-narrowing. Measuring FFR after stent placement was encouraged but not required. Intravascular imaging (using sound waves or light to see inside the artery) was used at the discretion of the treating physician. For patients assigned to CABG, surgery was performed according to standard practice at each center, with **complete arterial revascularization strongly recommended**. FFR testing to guide CABG was not required, but if it had been performed during the diagnostic angiogram, the surgeon could use that information. All patients in both groups received aspirin, a high-dose statin (cholesterol-lowering medication), and guideline-directed medical therapy for their heart disease. Patients who underwent PCI were required to take a second antiplatelet medication (to prevent blood clots) for at least 6 months after the procedure. Patients were followed at hospital discharge and then at 1, 6, and 12 months after their procedure. **The primary end point** — the main outcome the study was measuring — was the occurrence within 1 year of a **major adverse cardiac or cerebrovascular event**, defined as: - Death from any cause - Myocardial infarction (heart attack) - Stroke - Repeat revascularization (needing another procedure to restore blood flow) Heart attacks were classified as either procedural (related to the procedure itself) or spontaneous (occurring on their own later). For procedural heart attacks after stenting or surgery, the definition required a rise in the cardiac troponin level (a protein released when heart muscle is damaged) to **more than 10 times the 99th percentile of the upper reference limit** within 72 hours after the procedure, plus at least one additional finding: new abnormal Q waves on the electrocardiogram, a new left bundle-branch block, proof of a new blockage in a graft or major native coronary artery, or imaging evidence of new loss of viable heart muscle or a new regional wall-motion abnormality. This definition aligns with a **type 5 myocardial infarction** (post-bypass surgery heart attack) according to the Third and Fourth Universal Definitions of Myocardial Infarction. Spontaneous heart attacks were defined as a rise or fall in troponin with at least one value above the 99th percentile upper reference limit, together with evidence of myocardial ischemia (inadequate blood flow to heart muscle), such as symptoms of ischemia, electrocardiographic changes, new Q waves, or imaging evidence of new loss of viable muscle. An independent clinical events committee reviewed and adjudicated (officially judged) all events in a blinded fashion — meaning they did not know which treatment each patient had received. ### Understanding the Statistical Approach This was a **noninferiority trial**, which means the goal was to determine whether FFR-guided PCI is "not worse than" CABG by more than a certain pre-specified amount. This is different from a standard "superiority" trial, which asks whether one treatment is actually better than the other. Initially, the researchers estimated that 12% of CABG patients would have a primary end-point event within 1 year, based on previous trials. They originally planned to declare noninferiority if the **upper boundary of the 95% confidence interval was less than 1.45** for the hazard ratio. With that criterion, they calculated they would need 712 patients per group (1,424 total) to achieve 90% power. However, during recruitment and without knowledge of event rates, the trial's steering committee decided to change the noninferiority margin to an upper boundary of **less than 1.65**. This decision was based on newly published trials comparing CABG with PCI, which reported major adverse cardiac or cerebrovascular events in no more than 10% of CABG patients and used similar noninferiority margins. With the wider margin, only 645 patients per group (1,290 total) would be needed for 90% power. Nevertheless, the committee chose to complete the planned enrollment of 1,500 patients. The primary analysis was conducted on an **intention-to-treat basis**, meaning patients were analyzed in the group to which they were randomly assigned, regardless of what treatment they actually received. This approach preserves the benefits of randomization and reflects real-world practice. ## Who Took Part in the Study A total of **1,500 patients** were enrolled at 48 centers and underwent randomization. Of these, **757 were assigned to receive FFR-guided PCI** and **743 were assigned to receive CABG**. Because of randomization, the two groups were very similar in their characteristics. Here is a detailed breakdown of the patients in each group: - **Age:** 65.2 ± 8.6 years in the PCI group vs. 65.1 ± 8.3 years in the CABG group (mean age overall: 65 years) - **Male sex:** 81.4% (616 patients) in the PCI group vs. 83.3% (619 patients) in the CABG group - **White race:** 93.9% (711 patients) vs. 92.3% (686 patients) - **Body-mass index (BMI):** 28.6 ± 4.5 vs. 28.7 ± 4.3 (both groups were in the overweight range) - **Diabetes:** 28.3% (214 patients) vs. 28.8% (214 patients) — overall, 29% of all study participants had diabetes - Insulin-dependent diabetes: 7.3% (55 patients) vs. 8.2% (61 patients) - Non-insulin-dependent diabetes: 21.0% (159 patients) vs. 20.6% (153 patients) - **Hypertension:** 71.2% vs. 75.0% - **Dyslipidemia (abnormal cholesterol):** 68.9% vs. 71.7% - **Current tobacco use:** 19.2% vs. 18.4% - **Previous tobacco use:** 39.2% vs. 39.9% - **Family history of coronary artery disease:** 32.5% vs. 28.8% - **Previous heart attack (myocardial infarction):** 33.3% vs. 33.5% - **Previous PCI:** 13.0% vs. 14.0% - **History of transient ischemic attack (TIA) or stroke:** 6.5% vs. 7.6% - **Kidney disease:** 4.9% vs. 5.9% (defined as an estimated glomerular filtration rate below 60 ml per minute per 1.73 m²) - **Positive noninvasive test for ischemia:** 41.1% vs. 40.6% - **Left ventricular ejection fraction (LVEF) ≤ 50%:** 18.2% vs. 17.6% — meaning most patients had relatively preserved heart-pumping function - **Hospitalized with non-ST-segment elevation acute coronary syndrome (NSTE-ACS):** 39.7% vs. 38.7% — meaning about 39% of patients in each group had been admitted to the hospital for unstable angina or a milder form of heart attack The severity of the patients' coronary disease was also similar between groups. On average, patients had **4.3 blockages** in their heart arteries. About 22% of patients had at least one artery that was completely blocked (a **chronic total occlusion**), and 68% had at least one **bifurcation lesion** (a blockage at a point where an artery divides into two branches). The average **SYNTAX score** — an angiography-based scoring system that evaluates the severity and complexity of coronary artery disease (the lowest score is 0 and there is no upper limit; lower scores indicate less complex disease and predict better outcomes with PCI) — was 26 in both groups, indicating a moderate-to-high level of disease complexity. ## What Treatments Patients Actually Received The procedures the patients received differed substantially between the two groups, as expected. ### Timing and Hospital Stay - **Time to procedure:** The median time from randomization to the procedure was 4 days (interquartile range 1–13 days) for PCI vs. 13 days (6–26 days) for CABG. - **Procedure duration:** The median procedure time was 87 minutes (67–113 minutes) for PCI vs. 197 minutes (155–239 minutes) for CABG. - **Hospital stay:** The median length of hospitalization was 3 days (1–7 days) for PCI vs. 11 days (7–16 days) for CABG. ### Details of the PCI Procedures Among the 757 patients assigned to FFR-guided PCI, the average number of blockages treated per patient was 4.3, and a mean of **3.7 ± 1.9 stents** were implanted per patient. The median total length of stents placed was **80 mm** (52–116 mm). 22.1% of patients (166 of 750) had the procedure done in stages (more than one session). FFR was successfully measured in **82% of all blockages**. The most common reason FFR was not measured was that the artery was subtotally or completely blocked — in these cases, the blockage was clearly significant and FFR measurement would not add useful information. The mean FFR value was **0.70**, confirming that most treated blockages were indeed severely limiting blood flow. Importantly, **24% of the blockages that were originally intended for treatment had an FFR greater than 0.80**, meaning they were not actually limiting blood flow. Under the study protocol, these blockages were **not stented** — demonstrating how FFR guidance prevents unnecessary procedures. After stenting, FFR was re-measured in 60% of treated blockages, with a mean value of **0.88**, confirming that stent placement successfully restored blood flow. Intravascular imaging was used in only **11.7% of cases** (87 of 744 patients) — meaning most stent placements were guided by angiography and FFR alone. ### Details of the CABG Procedures Among the 743 patients assigned to CABG, the average number of blockages was 4.2, and patients received a mean of **3.4 ± 1.0 distal anastomoses** (the surgical connections that create the bypass pathways). Notably: - **97.0%** of patients (684 of 705) received a **left internal thoracic artery graft** — the key arterial graft that provides the most durable bypass. - **24.5%** of patients (173 of 705) received **multiple arterial grafts** — meaning more than one artery was used as a bypass vessel. The study protocol strongly recommended complete arterial revascularization, but this was achieved in only about a quarter of patients. - **24.1%** of operations (168 of 698) were performed **off-pump** — meaning without the use of a heart-lung bypass machine. - FFR was measured before CABG in only **10.0%** of patients (72 of 718), since FFR guidance for bypass surgery is not standard practice. ## Key Findings: The Primary Result Follow-up at 1 year was completed in **99.7% of patients** — an exceptionally high rate that makes the results highly reliable. The study's central finding is clear: **FFR-guided PCI did NOT meet the criteria for noninferiority to CABG.** In other words, the study could not confirm that stenting was "not worse than" bypass surgery for patients with three-vessel coronary artery disease. At 1 year, the incidence of the primary end point (death from any cause, heart attack, stroke, or repeat revascularization) was: - **10.6%** in the FFR-guided PCI group (80 of 757 patients) - **6.9%** in the CABG group (51 of 743 patients) - **Hazard ratio: 1.5** (95% confidence interval [CI], 1.1 to 2.2) - **P = 0.35 for noninferiority** — meaning the results did not meet the statistical requirement to declare PCI "not worse than" CABG To understand why this failed, it helps to understand the numbers. A **hazard ratio of 1.5** means that patients who received PCI had a 50% higher risk of experiencing a major adverse event during the year compared with patients who received CABG. The 95% confidence interval (1.1 to 2.2) provides a range of plausible values for the true risk; since **both the lower bound (1.1) and the upper bound (2.2) are above 1.0**, the difference between the groups was statistically significant in favor of CABG. Moreover, the upper bound of 2.2 exceeded the pre-specified noninferiority threshold of 1.65, meaning the results were not even close to declaring PCI "not worse" — they actually suggested CABG was better. The study's original assumption that 12% of CABG patients would have an event within 1 year turned out to be too pessimistic — the actual rate was 6.9%, meaning CABG performed even better than expected. ## Detailed Results: Individual Events When the researchers broke down the primary end point into its individual components, no single outcome was dramatically different between the two groups on its own, but the pattern consistently favored CABG: - **Death from any cause:** 1.6% (12 patients) in the PCI group vs. 0.9% (7 patients) in the CABG group (hazard ratio, 1.7; 95% CI, 0.7 to 4.3) - **Death from cardiac causes:** 0.8% (6 patients) vs. 0.5% (4 patients) - **Myocardial infarction (heart attack):** 5.2% (39 patients) vs. 3.5% (26 patients) (hazard ratio, 1.5; 95% CI, 0.9 to 2.5) - Spontaneous heart attacks: 3.3% (25 patients) vs. 2.3% (17 patients) - Procedural heart attacks: 1.7% (13 patients) vs. 1.2% (9 patients) - **Stroke:** 0.9% (7 patients) vs. 1.1% (8 patients) (hazard ratio, 0.9; 95% CI, 0.3 to 2.4) — this was the only component where the two treatments came close to equal - **Repeat revascularization:** 5.9% (45 patients) vs. 3.9% (29 patients) (hazard ratio, 1.5; 95% CI, 0.9 to 2.3) - Repeat PCI: 5.2% (39 patients) vs. 3.5% (26 patients) - Repeat CABG: 0.8% (6 patients) vs. 0.4% (3 patients) The composite of **death, heart attack, or stroke** (excluding repeat revascularization) occurred in **7.3%** of PCI patients (55 patients) vs. **5.2%** of CABG patients (39 patients) — a hazard ratio of 1.4 (95% CI, 0.9 to 2.1). While this difference was not statistically significant on its own, it followed the same direction as the overall result. An important note: the confidence intervals for these secondary end points were not adjusted for multiple comparisons, so they should not be interpreted as definitive evidence of treatment effects for individual outcomes. However, the consistency of the pattern is telling. At 1 year, there were no obvious differences between the groups regarding medical therapy, with one exception: a higher percentage of patients in the FFR-guided PCI group were receiving dual antiplatelet therapy and nitrate medication. ## Safety: Short-Term Complications While CABG was more effective at preventing major cardiac events at 1 year, the **safety profile** of the two procedures was quite different. Bypass surgery is a major operation, and it comes with substantially higher risks of certain complications in the early post-procedure period: - **Major bleeding (BARC type 3–5):** 1.6% (12 patients) in the PCI group vs. **3.8%** (28 patients) in the CABG group (P = 0.009) — meaning major bleeding was more than twice as common after bypass surgery - **Acute kidney injury:** 0.1% (1 patient) vs. **0.9%** (7 patients) (P = 0.04) — defined as a rise in serum creatinine of at least 0.3 mg per deciliter within 48 hours, or a 1.5-fold increase from baseline, or reduced urine output - **Atrial fibrillation or clinically significant arrhythmia:** 2.4% (18 patients) vs. **14.1%** (105 patients) (P < 0.001) — irregular heart rhythms were nearly six times more common after bypass surgery - **Rehospitalization within 30 days:** 5.5% (42 patients) vs. **10.2%** (76 patients) (P < 0.001) — patients who had bypass surgery were nearly twice as likely to be readmitted to the hospital within a month There were also procedure-specific complications unique to each approach: - **Definite stent thrombosis** (a blood clot forming inside a newly placed stent) occurred in **0.8%** of PCI patients (6 patients). - **Definite symptomatic graft occlusion** (a bypass graft becoming blocked) occurred in **1.3%** of CABG patients (10 patients). These safety data highlight the trade-off that lies at the heart of this study. Bypass surgery may provide more durable, complete revascularization — leading to fewer heart attacks, deaths, and repeat procedures at 1 year — but it comes at the cost of a more invasive procedure with significantly higher rates of short-term complications like bleeding, kidney injury, arrhythmia, and hospital readmission. ## Subgroup Analyses: Who Benefits From Which? The researchers also examined whether the results were consistent across different types of patients. They pre-specified the following subgroup categories to analyze: - Age group: under 65 years vs. 65 years or older - Sex: male vs. female - Presence or absence of diabetes - Presence or absence of acute coronary syndrome at presentation - Left ventricular ejection fraction: above 50% vs. 50% or below - Previous PCI: yes vs. no - SYNTAX score category: low (0–22), intermediate (23–32), or high (>32) The results of these subgroup analyses were displayed in a forest plot, and they showed that the treatment effect was **generally consistent across all subgroups**. That is, the pattern of CABG being associated with fewer events than PCI held across different age groups, in men and women, in patients with and without diabetes, across different levels of heart function, and regardless of previous PCI history. No subgroup was found in which PCI clearly matched or surpassed CABG's results. This consistency ## Frequently Asked Questions ### What is three-vessel coronary artery disease? Three-vessel coronary artery disease means blockages develop in all three major heart arteries, seriously reducing blood supply to the heart muscle. Without treatment, patients face a high risk of heart attacks, heart failure, and death. This condition was the focus of the FAME 3 study, which compared two treatment approaches. ### What are the two treatment options for three-vessel heart disease? The two options are coronary-artery bypass grafting (CABG), which is open-heart surgery that creates bypass pathways around blockages, and percutaneous coronary intervention (PCI), commonly called stenting, which uses a catheter to place a mesh tube inside the artery. The FAME 3 trial compared these approaches in 1,500 patients. ### What is fractional flow reserve (FFR) and why does it matter? FFR is a technique used during stenting to measure blood pressure before and after a blockage using a thin wire. A value of 0.80 or lower means the blockage limits blood flow and deserves a stent. FFR guidance prevents unnecessary stenting of blockages that are not actually restricting blood flow. ### What were the main results of the FAME 3 study? At one year, 10.6% of patients who received FFR-guided stents had a major heart-related event (death, heart attack, stroke, or repeat procedure) compared with 6.9% who had bypass surgery. The stent procedure did not meet the study's criteria to be considered not worse than surgery. ### What short-term complications were more common with bypass surgery? Bypass surgery had significantly higher rates of major bleeding, acute kidney injury, atrial fibrillation or irregular heart rhythms, and rehospitalization within 30 days. For example, major bleeding occurred in 3.8% of surgery patients versus 1.6% of stent patients, and irregular rhythms in 14.1% versus 2.4%. ### Who was included in the FAME 3 study? The study included 1,500 patients with blockages in all three major heart arteries, without left main disease. Patients were randomly assigned to stenting or bypass surgery. About 29% had diabetes, average age was 65 years, and most were male. Patients with recent severe heart attacks or severely weakened pumping function were excluded. --- Publisher: Diagnostic Detectives Network (https://diagnosticdetectives.com) — independent multi-expert medical second opinions, worldwide, private-pay. Author byline: Anton Titov, MD, PhD. Contact: https://diagnosticdetectives.com/pages/contact Canonical page: https://diagnosticdetectives.com/products/stents-vs-bypass-surgery-for-three-vessel-heart-disease-what-the-fame-3-study-found