Online Supplement to Nudi et al, Prognostic impact of location and extent of vessel-related ischemia at myocardial perfusion scintigraphy in patients with or at risk for coronary artery disease
1
Online Table 1. Clinical outcomes according to location and extent of vessel-related ischemia (VRI) at myocardial perfusion scintigraphy.
No VRI (N=5,436) / Minimal or mild VRI (N=5,191) / Moderate or severe VRI / Overall p / Other p <0.05*Single-VRI / Multi-VRI
Involving LAD (N=749) / Not involving LAD (N=878) / Involving LAD (N=428) / Not involving LAD (N=572)
Follow-up duration (months) / 34±21 / 30±21 / 32±22 / 32±22 / 31±21 / 31±21 / <0.001 / -
Death or myocardial infarction / 100 (1.8%) / 144 (2.8%) / 36 (4.8%) / 40 (4.6%) / 35 (8.2%) / 29 (5.1%) / <0.001 / b,d,f
Cardiac death or myocardial infarction / 75 (1.4%) / 92 (1.8%) / 31 (4.1%) / 32 (3.6%) / 28 (6.5)% / 24 (4.2%) / <0.001 / d
Death / 42 (0.8%) / 92 (1.8%) / 20 (2.7%) / 29 (3.3%) / 27 (6.3%) / 17 (3.0%) / <0.001 / b,d,f
Cardiac death / 17 (0.3%) / 40 (0.8%) / 15 (2.0%) / 21 (2.4%) / 20 (4.7%) / 12 (2.1%) / <0.001 / b,d,f
Non-fatal myocardial infarction / 58 (1.1%) / 52 (1.0%) / 16 (2.1%) / 11 (1.3%) / 8 (1.9%) / 12 (2.1%) / 0.005 / -
Percutaneous coronary intervention / 231 (4.2%) / 655 (12.6%) / 325 (43.4%) / 291 (33.1%) / 144 (33.6%) / 239 (41.8%) / <0.001 / a,b,e,f
Coronary artery bypass grafting / 29 (0.5%) / 162 (3.1%) / 184 (24.6%) / 75 (8.5%) / 152 (35.5%) / 74 (12.9%) / <0.001 / a,b,c,d,e
Any revascularization / 256 (4.7%) / 798 (15.4%) / 490 (65.4%) / 357 (40.7%) / 287 (67.1%) / 310 (54.2%) / <0.001 / a,c,d,e,f
*a=Single-VRI – involving LAD vs single-VRI - not involving LAD; b=Single-VRI – involving LAD vs multi-VRI – involving LAD; c=Single-VRI – involving LAD vs multi-VRI – not involving LAD; d=Single-VRI – not involving LAD vs multi-VRI – involving LAD; e=Single-VRI – not involving LAD vs multi-VRI – not involving LAD; f=multi-VRI – involving LAD vs multi-VRI – not involving LAD; LAD=left anterior descending
Online Table 2. Multivariable analysis exploring the independent long-term prognostic impact of moderate or severe vessel-related ischemia (VRI) at myocardial perfusion scintigraphy (MPS).*
Outcome / VRIDeath or myocardial infarction / HR=1.17 (1.04-1.08), p=0.010
Cardiac death or myocardial infarction / HR=1.14 (1.05-1.24), p=0.003
Death / HR=1.16 (1.01-1.34), p=0.043
Cardiac death / HR=1.06 (0.88-1.27), p=0.532
Non-fatal myocardial infarction / HR=1.16 (1.04-1.30), p=0.006
*including in the Cox proportional hazard model with backwise stepward selection (p for exit 0.10) age, gender, hypercholesterolemia, hypertriglyceridemia, smoking, diabetes mellitus, revascularization before MPS, type of stress, workload, rate pressure product, ST-segment changes, ejection fraction, end-diastolic volume, summed stress score, and VRI (distinguished as 0-no ischemia, 1-mild or moderate VRI, 2-single-VRI – not involving LAD, 3-multi-VRI – not involving LAD, 4-single-VRI – involving LAD, 5-multi-VRI – involving LAD), and reported as hazard ratios (HR) with 95% confidence intervals
Online Table 3. Significant predictors at incremental multivariable Cox proportional hazard analysis for long-term (>1 year) death or myocardial infarction.*
Model 1(clinical data) / Model 2
(model 1 plus stress test data) / Model 3
(model 2 plus EF and EDV) / Model 4
(model 3 plus SSS) / Model 5
(model 4 plus SSS)
Age
HR=1.08 (1.06-1.09), p<0.001
Female gender
HR=0.66 (0.51-0.85), p=0.001
Diabetes mellitus
HR=1.28 (0.98-1.54), p=0.024
Prior revascularization HR=0.79 (0.64-0.98), p=0.035 / Age
HR=1.04 (1.02-1.06), p<0.001
Female gender
HR=0.60 (0.40-0.90), p=0.013
RPP
HR=0.94 (0.90-0.98), p=0.003
ST deviation
HR=1.55 (1.14-2.11), p=0.005 / Age
HR=1.05 (1.06-1.98), p<0.001
RPP
HR=0.95 (0.91-0.99), p=0.011
ST deviation
HR=1.45 (1.06-1.98), p=0.019
EF
HR=0.97 (0.96-0.98), p<0.001 / Age
HR=1.05 (1.03-1.07), p<0.001
EF
HR=0.97 (0.96-0.99), p<0.001
SSS
HR=1.46 (1.30-1.64), p<0.001 / Age
HR=1.07 (1.06-1.08), p<0.001
EF
HR=0.96 (0.95-0.97), p<0.001
VRI
HR=1.17 (1.04-1.08), p=0.010
* with backwise stepward selection (p for exit 0.10) reported as hazard ratio (HR) with 95% confidence interval and corresponding p value; model 1 included age, gender, hypercholesterolemia, hypertriglyceridemia, smoking, diabetes mellitus, revascularization before MPS; model 2 included variables from model 1 plus type of stress, workload, rate pressure product (RPP, divided by 1,000), ST-segment changes (≥1.0 mm); model 3 included variables in model 2 plus ejection fraction (EF) and end-diastolic volume (EDV); model 4 included variables from model 3 plus and summed stress score (SSS); model 5 included variables from model 4 plus vessel-related ischemia (VRI) as coded in Online Table 2
Online Table 4. Contingency table showing the breakdown of vessel-related ischemia (VRI) at myocardial perfusion scintigraphy (MPS) according to coronary artery disease at invasive angiography performed after MPS.
Single-VRI / Multi-VRI
Involving LAD (N=386) / Not involving LAD (N=304) / Involving LAD (N=210) / Not involving LAD (N=261)
Coronary angiography / <50% diameter stenosis / 160 (13.6%) / 8 (2.1%) / 4 (1.3%) / 6 (2.9%) / 1 (0.4%)
≥50% diameter stenosis / Single-vessel disease / LAD territory / 314 (26.7%) / 165 (42.7%) / 46 (15.1%) / 16 (7.6%) / 34 (13.0%)
Not involving LAD / 67 (5.7%) / 2 (0.5%) / 33 (10.9%) / 4 (1.9%) / 10 (3.8%)
Multi-vessel disease / Involving LAD / 575 (48.9%) / 207 (53.6%) / 187 (61.5%) / 178 (84.8%) / 47 (18.0%)
Not involving LAD / 61 (5.2%) / 4 (1.0%) / 34 (11.2%) / 6 (2.9%) / 169 (64.8%)
LAD=left anterior descending
Online Figure 1. Explicative image to highlight how the maximal ischemia score (MIS) and the vessel-related ischemia (VRI) concepts can be combined to obtain a more prognostically relevant Final Score (FS), with the number inside squares representing the combination of these two instruments with color codes proportional to the severity of the FS. LAD=left anterior descending.
Online Figure 2. Kaplan-Meier failure curve for the occurrence of cardiac death or myocardial infarction, distinguishing patients according to vessel-related ischemia (VRI), and excluding patients undergoing revascularization as first follow-up event (overall log-rank p<0.001; other log-rank p<0.05: b,d,f [see legend of Table 1 for explanations]). Color codes: blue - no VRI; brown - minimal or mild VRI; green – single-VRI – involving LAD; dark orange – single-VRI - not involving LAD; gray – multi-VRI - involving LAD; red – multi-VRI - not involving LAD.
Online Figure 3. Kaplan-Meier failure curve for the occurrence of death, distinguishing patients according to vessel-related ischemia (VRI), and excluding patients undergoing revascularization as first follow-up event (overall log-rank p<0.001; other log-rank p<0.05: b,d,f [see legend of Table 1 for explanations]). Color codes: blue - no VRI; brown - minimal or mild VRI; green – single-VRI – involving LAD; dark orange – single-VRI - not involving LAD; gray – multi-VRI - involving LAD; red – multi-VRI - not involving LAD.
Online Figure 4. Kaplan-Meier failure curve for the occurrence of cardiac death, distinguishing patients according to vessel-related ischemia (VRI), and excluding patients undergoing revascularization as first follow-up event (overall log-rank p<0.001; other log-rank p<0.05: b,d,f [see legend of Table 1 for explanations]). Color codes: blue - no VRI; brown - minimal or mild VRI; green – single-VRI – involving LAD; dark orange – single-VRI - not involving LAD; gray – multi-VRI - involving LAD; red – multi-VRI - not involving LAD.
Online Figure 5. Kaplan-Meier failure curve for the occurrence of myocardial infarction, distinguishing patients according to vessel-related ischemia (VRI), and excluding patients undergoing revascularization as first follow-up event (overall log-rank p<0.001; other log-rank p>0.05). Color codes: blue - no VRI; brown - minimal or mild VRI; green – single-VRI – involving LAD; dark orange – single-VRI - not involving LAD; gray – multi-VRI - involving LAD; red – multi-VRI - not involving LAD.
1