Methods

We simulated the catheterization laboratory management of a cohort of actual patients with NSTEMIs in whom pressure-wire studies had been performed because of diagnostic uncertainty on the basis of visual assessment of the coronary angiogram. The study was approved by the Clinical Governance Department of Golden Jubilee National Hospital.

Consecutive patients with NSTEMIs who had undergone clinically indicated coronary angiography with FFR measurement from January 2009 to March 2010 were included. Golden Jubilee National Hospital is a regional cardiothoracic center with 12 interventional cardiologists. In our hospital, treatment decisions for medical therapy, percutaneous coronary intervention (PCI; including bare-metal and drug-eluting stents), coronary artery bypass grafting (CABG), and deferred management in patients with NSTEMI are made according to contemporary guidelines.

Patients were identified from the hospital’s catheterization laboratory registry, which prospectively accumulates the clinical data of all patients who undergo invasive management. The inclusion criteria were (1) a diagnosis of recent (≤5 days) NSTEMI and (2) an FFR measurement intended to guide diagnostic management during coronary angiography. The only exclusion criterion was an FFR measurement for other reasons (e.g., FFR measured after stent deployment). The definition of a culprit lesion was a stenosis in a coronary artery causally implicated in the NSTEMI, on the basis of angiographic appearances (e.g., eccentric, hazy appearance) and regional ischemic electrocardiographic changes or wall motion abnormalities on echocardiography. The definition of a nonculprit stenosis was a lesion >30% of the reference vessel diameter not implicated in the acute presentation.

The indication for FFR measurement was the presence of an intermediate coronary lesion (i.e., 40% to 80% stenosis severity) associated with diagnostic and treatment uncertainty. FFR measurement was used to provide functional information on lesion severity, and FFR ≤0.80 was taken to represent a flow-limiting stenosis. An FFR threshold of 0.80 identifies ischemia-causing coronary stenoses with high diagnostic accuracy (>90% sensitivity and specificity).

FFR measurements were obtained according to standard practice in our hospital using a 0.014-inch pressure-sensitive wire (RADI; St. Jude Medical Corporation, Uppsala, Sweden) during intravenous adenosine infusion (140 μg/kg/min) to establish maximal coronary hyperemia and confirmed with typical changes in blood pressure and symptoms.

The study simulated the actual clinical decision-making process by taking place in the cardiac catheterization laboratory (September to November 2010). Five accredited interventional cardiologists (A, B, C, D, and E) independently reviewed each case of NSTEMI, and a sixth cardiologist acted as the study coordinator.

These cardiologists had a broad range of postaccreditation experience: 1 year (cardiologist A), 2 years (cardiologist B), 5 years (cardiologist C), 15 years (cardiologist E), and 30 years (cardiologist D). Three of the cardiologists had worked in the United States, and 4 of the cardiologists had worked in Canada. Each cardiologist performed ≥250 PCIs per year.

Each interventional cardiologist met with the coordinator separately. The cardiologists were blinded to patient identity and the original treatment decisions ( Table 1 ). Each patient’s clinical records and coronary angiogram were reviewed, and on the basis of this information, the cardiologist then made a management decision. The FFR results were then disclosed, and the initial management decision was reviewed in light of the FFR result and changed (or not), as appropriate.

The initial treatment decisions of the 5 cardiologists (A, B, C, D, and E) on the basis of visual assessment of the angiogram alone were assessed using a chi-square test. Comparisons among the 5 cardiologists for each treatment option before and after disclosure of the FFR results were assessed using Fisher’s exact test. Consensus (or majority) treatment decisions were defined as ≥3 of the 5 cardiologists’ agreeing on the same treatment decision.

For our third aim, we studied the patient and operator characteristics that might influence compliance with an FFR result and propensity to change an initial treatment decision. We created a proportional odds model on 3 possible outcomes: (1) no change, (2) change to medical management, and (3) change to revascularization with either PCI or CABG with rater (i.e., the cardiologist) as a random effect and an independent covariance matrix. This approach yielded univariate odds models for the following characteristics: each cardiologist (rater), gender, clinical significance of the lesion (culprit, nonculprit, or not determinable), diabetes, previous CABG, previous myocardial infarction, previous PCI, and age. All of these models were adjusted for rater. A multivariate odds model was created adjusting simultaneously for all these characteristics, regardless of statistical significance. The odds ratio is the cumulative odds ratio (outcome 2 vs outcome 1, i.e., change to medical management vs no change, and then outcome 3 vs outcomes 1 and 2, i.e., change to revascularization with either PCI or CABG vs change to medical management or no change). An odds ratio >1 indicates a higher probability of a subject with that characteristic making the specified change.

All p values <0.05 were taken as significant, and no adjustment was made for multiple comparisons. Statistical analysis was performed using SAS version 9.2 (SAS Institute Inc., Cary, North Carolina).

Results

FFR was measured in 109 patients with NSTEMIs (January 1, 2009, to March 31, 2010). Nine of these patients were excluded because FFR was measured after stenting (n = 8) or because FFR measurement failed because of vessel tortuosity (n = 1). Therefore, 100 different patients with NSTEMIs were included in the study, and their clinical characteristics and actual treatments are listed in Table 1 . Of all PCI procedures (n = 66), 33 did not involve FFR. Representative cases are shown in Figures 1 and 2 .

Left coronary angiogram of a 44-year-old man who presented with an acute NSTEMI and anterior ST-segment changes on electrocardiography. The angiogram, obtained 3 days after admission to the hospital, revealed an intermediate stenosis of equivocal severity in the distal half of the left main stem coronary artery, which was suspected to be the culprit vessel. The right coronary angiogram (not shown) revealed a 70% stenosis distally. A pressure-wire study was performed in the left main (orange arrows) , and the FFR value was 0.70, indicating a flow-limiting stenosis. The patient was subsequently discussed at the multidisciplinary team meeting, and the decision was for coronary bypass. He has since progressed well over the past year. LAO = left anterior oblique; RAO = right anterior oblique.

Left coronary angiogram of a 71-year-old man with diabetes obtained 1 day after hospital admission with an acute NSTEMI associated with lateral ST-segment changes on electrocardiography. The angiogram revealed a culprit stenosis of the obtuse marginal coronary artery (yellow arrows) and potentially obstructive stenoses in the proximal and mid parts of the left anterior descending coronary artery (LAD). Because there was diagnostic uncertainty, a pressure-wire study was performed in the LAD, and the FFR across the left main–mid segment of the LAD (orange arrow) was 0.94, indicating a non-flow-limiting stenosis. The patient was then treated with a single drug-eluting stent in the obtuse marginal artery and discharged from the hospital later that day. He has since progressed well. LAO = left anterior oblique; RAO = right anterior oblique.

The initial treatment decisions on the basis of clinical data alone differed overall among the 5 cardiologists (p = 0.0061).

Changes in treatment decisions after FFR disclosure are listed in Tables 2 to 4 . The number of patients in whom the treatment decisions made by each cardiologist independently conformed (and so represented a majority with ≥3 of the 5 cardiologists) increased from 65% on the basis of coronary angiography alone to 91% after FFR disclosure (p = 0.0094; Table 4 ).

Table 3

Changes from the initial treatment plan after fractional flow reserve disclosure according to the specific types of treatments (medical therapy, single-vessel percutaneous coronary intervention, multivessel percutaneous coronary intervention, or coronary artery bypass grafting)

Type of Change	Direction of Change	Cardiologist (Rater)					Average Change	Rater Effect p Value
		A	B	C	D	E
No change		48%	51%	53%	51%	69%	54%
Any change		52%	49%	47%	49%	31%	46%	0.0016
Medical management	To	30%	23%	29%	22%	16%	24%	0.0064
	From	4%	4%	0%	2%	3%	3%	NA
Single vessel PCI	To	16%	18%	13%	18%	14%	16%	0.68
	From	22%	21%	14%	14%	8%	16%	0.0052
Multivessel PCI	To	5%	5%	3%	5%	1%	4%	NA
	From	10%	12%	8%	10%	4%	9%	0.046
CABG	To	1%	3%	2%	4%	0%	2%	NA
	From	0%	4%	1%	0%	3%	2%	NA
Deferred management	To	0%	0%	0%	0%	0%	0.0%	NA
	From	16%	8%	24%	23%	12%	17%	0.0067

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Safety and Efficacy of Everolimus -Eluting Stents Versus Sirolimus -Eluting Stents in Women
2. Analysis of the Prevalence of Cardiovascular Disease and Associated Risk Factors for European-American and African-American Populations in the State of Pennsylvania 2005–2009
3. Comparison of Variables in Men Versus Women Undergoing Transcatheter Aortic Valve Implantation for Severe Aortic Stenosis (from Italian Multicenter CoreValve Registry)
4. Erratum for Scott M. Grundy, et al. “Cardiorespiratory Fitness and Metabolic Risk.” Am J Cardiol 2012; 109:988–993

Stay updated, free articles. Join our Telegram channel

Join