Meta-Analysis of Randomized Studies Comparing Intravascular Ultrasound Versus Angiographic Guidance of Percutaneous Coronary Intervention in Pre–Drug-Eluting Stent Era




We conducted a formal meta-analysis of peer-reviewed, published, randomized studies comparing intravascular ultrasound (IVUS)-guidance and angiographic-guided bare metal stent implantation. A total of 8 studies were identified. Because the Balloon Equivalent to Stent (BEST) study was a noninferiority trial designed to compare 2 very different percutaneous coronary intervention strategies—IVUS-guided aggressive balloon angioplasty (with bail-out stenting) and angiographic-guided deliberate bare metal stent implantation—it was eliminated. An unadjusted random-effects meta-analysis was used to compare the IVUS-guided and non–IVUS-guided stenting in the 7 remaining studies. A total of 2,193 patients were randomized in 5 multicenter and 2 single-center studies. IVUS guidance was associated with a significantly larger postprocedure angiographic minimum lumen diameter. The mean difference was 0.12 mm (95% confidence interval [CI] 0.06 to 0.18, p <0.0001). IVUS guidance was also associated with a significantly lower rate of 6-month angiographic restenosis (22% vs 29%, odds ratio 0.64, 95% CI 0.42 to 0.96, p = 0.02), a significant reduction in the revascularization rate (13% vs 18%, odds ratio 0.66, 95% CI 0.48 to 0.91, p = 0.004), and overall major adverse cardiac events (19% vs. 23%, odds ratio 0.69, 95% CI 0.49 to 0.97, p = 0.03). However, no significant effect was seen for myocardial infarction (p = 0.51) or mortality (p = 0.18). In conclusion, IVUS guidance for bare metal stent implantation improved the acute procedural results (angiographic minimum lumen diameter) and thereby reduced angiographic restenosis and repeat revascularization and major adverse cardiac events, with a neutral effect on death and myocardial infarction during a follow-up period of 6 months to 2.5 years.


The intravascular ultrasound (IVUS) observations of Nakamura et al and Colombo et al fundamentally changed the methods used to implant stents, leading to randomized trials to assess the effect of IVUS guidance on stent implantation, often with conflicting results. Systematic reviews and meta-analyses have integrated the results of independent studies considered combinable to allow an objective appraisal of the evidence, provide a more precise estimate of the treatment effect, and explain the heterogeneity among individual studies. Since the publication of a previous meta-analysis, the Direct Stenting vs Optimal angioplasty (DIPOL) study, Angiography vs Intravascular Ultrasound-Directed (AVID) study, the Gaster study, and a long-term follow-up of the Restenosis after IVUS-Guided Stenting (RESIST) study have been published. This has brought to a close the era of randomized studies comparing IVUS and angiographic guidance of percutaneous coronary intervention (PCI) with bare metal stents (BMSs). It has been estimated that 30% of stents implanted worldwide are BMSs and not drug-eluting stents. Therefore, we report a formal meta-analysis of all peer-reviewed, published, randomized studies comparing IVUS-guided and angiographic-guided PCI in the pre-drug-eluting stent era.


Methods


We performed a MEDLINE search to identify all published, English-language, peer-reviewed studies published through December 31, 2009 that had performed a randomized comparison of IVUS-guided versus angiographic-guided PCI and/or BMS implantation in which the ≥6-month angiographic and/or clinical follow-up data were reported. The most comprehensive results were obtained by searching the term “IVUS+angiography” (n = 940 before 2010), manually reviewing the titles, searching the references cited in the original and review reports, and using the Citations Index to identify newer studies that had cited older publications. Nonrandomized studies and those reporting only acute results were excluded. We found 8 published, randomized IVUS vs angiographic studies. Two investigators (G.S.M. and H.P.) reviewed the studies; disagreements were resolved by consensus. Because the Balloon Equivalent to Stent (BEST) study was a noninferiority trial that compared 2 very different PCI strategies—IVUS-guided aggressive balloon angioplasty (with bail-out stenting) and angiographic-guided deliberate BMS implantation —it was not included.


An unadjusted random-effects meta-analysis was used to compare IVUS- versus non–IVUS-guided stenting. The random effects model took into account the heterogeneity among the studies by assuming that the individual study effects were drawn from a random distribution. The test of heterogeneity using the Q statistic was computed, and the I 2 statistic was calculated to quantify the amount of heterogeneity. The results from the fixed effects model were computed for comparison. The binary variables were summarized as the frequencies or percentages, and continuous variables were summarized as the mean (±SD or range). For binary outcomes, the odds ratio (OR) was used for comparative purposes. For continuous variables, the treatment effect was assessed by way of the differences. The OR and 95% confidence intervals (CIs) were calculated for the clinical outcomes according to the intent to treat and for angiographic binary restenosis according to the patients who returned for angiographic follow-up. Forest plots of the OR for binary outcomes and of the mean differences for continuous variables were constructed, with the sizes of the boxes proportional to the number of patients in the individual studies. The null hypothesis of a nonzero treatment effect was examined using the directional test versus the alternative hypothesis that all effects would be equal to the same nonzero alternative. p Values <0.05 were considered statistically significant. To assess the potential publication bias, funnel plots were constructed by plotting the precision (the inverse of the standard error) against the log OR or the mean difference; and the Begg rank correlation test was computed.




Results


The baseline characteristics of the studies and patients are listed in Tables 1 and 2 , respectively, and were well matched in all 7 studies. A total of 2,193 patients were randomized in 5 multicenter and 2 single-center IVUS- versus angiographic-guided PCI studies.



Table 1

Study characteristics


























































Characteristic DIPOL AVID Gaster Study RESIST TULIP OPTICUS SIPS
Multicenter Yes Yes No Yes No Yes No
Total study size 163 800 108 155 150 548 269
Angiographic follow-up Yes No Yes Yes Yes Yes Yes
Duration (mo) 6 6 6 6 6 6
Total follow-up duration 6 mo 12 mo 2.5 yrs 18 mo 12 mo 12 mo 2 yrs

DIPOL study had 3 groups: group 1, direct stenting guided by quantitative coronary angiography; group 2, direct stenting guided by IVUS; and group 3, optimal balloon angioplasty guided by IVUS; the present analysis only compared groups 1 and 2.


Mean follow-up 2.5 years.



Table 2

Baseline patient data


































































































































































































































































































































































































































Variable DIPOL AVID Gaster RESIST TULIP OPTICUS SIPS
Patients (n)
Intravascular ultrasound guidance 83 406 54 79 74 273 121
Nonintravascular ultrasound guidance 80 394 54 76 76 275 148
Men
Intravascular ultrasound guidance 71% 73% 100% 86% 71% 77% 82%
Nonintravascular ultrasound guidance 73% 68% 100% 93% 72% 78% 76%
Age (years)
Intravascular ultrasound guidance 56 ± 8 62 ± 12 57 (40–73) 57 ± 10 61 ± 10 60.1 ± 10 61.2 ± 8.1
Nonintravascular ultrasound guidance 54 ± 8 63 ± 11 57 (33–78) 56 ± 12 63 ± 10 61.5 ± 9.5 60.7 ± 9.6
Ejection fraction (%)
Intravascular ultrasound guidance 52 ± 9 53 ± 13 65 ± 12 53 ± 13 NA 56.5 ± 14 NA
Nonintravascular ultrasound guidance 48 ± 10 55 ± 13 69 ± 12 51 ± 9 NA 57.7 ± 14.3 NA
Previous myocardial infarction
Intravascular ultrasound guidance 44% 35% 54% 68% NA 32% 58%
Nonintravascular ultrasound guidance 40% 29% 44% 63% NA 32% 52%
Smoker
Intravascular ultrasound guidance 47% NA 30% 55% 40% 69% 47%
Nonintravascular ultrasound guidance 52% NA 15% 51% 43% 66% 45%
Diabetes mellitus
Intravascular ultrasound guidance 10% 15% 4% 9% 16% 17% 16%
Nonintravascular ultrasound guidance 11% 17% 11% 8% 21% 17% 16%
Hypertension
Intravascular ultrasound guidance NA 46% 20% 24% 27% 48% 64%
Nonintravascular ultrasound guidance NA 45% 24% 26% 30% 52% 56%
Dyslipidemia
Intravascular ultrasound guidance 47% 40% 96% 54% 61% 61% 88%
Nonintravascular ultrasound guidance 40% 44% 93% 52% 62% 67% 87%
American College of Cardiology/American Heart Association B2/C lesions
Intravascular ultrasound guidance 13% NA 47% 43% 100% 76% 51%
Nonintravascular ultrasound guidance 10% NA 46% 48% 100% 78% 42%
Left anterior descending artery
Intravascular ultrasound guidance 41% 40% 48% 48% 39% 51% 38%
Nonintravascular ultrasound guidance 46% 37% 46% 47% 38% 50% 41%
Left circumflex artery
Intravascular ultrasound guidance 26% 15% 24% 11% 10% 18% 27%
Nonintravascular ultrasound guidance 24% 18% 26% 11% 21% 14% 27%
Right coronary artery
Intravascular ultrasound guidance 33% 35% 28% 41% 51% 30% 30%
Nonintravascular ultrasound guidance 30% 32% 28% 42% 41% 35% 30%
Saphenous vein graft
Intravascular ultrasound guidance 10% 5%
Nonintravascular ultrasound guidance 12% 3%
Left main coronary artery
Intravascular ultrasound guidance 0.8%
Nonintravascular ultrasound guidance 0.5%

Data presented as n, %, mean ± SD, or mean (range).

DIPOL study had 3 groups: group 1, direct stenting guided by quantitative coronary angiography; group 2, direct stenting guided by IVUS; and group 3, optimal balloon angioplasty guided by IVUS; the present analysis only compared groups 1 and 2.


History of smoking.


Current smoking.



All 7 studies included data on the postprocedure angiographic minimum lumen diameter ( Table 3 ). As shown in Figure 1 , IVUS guidance was associated with a significantly larger postprocedure angiographic minimum lumen diameter; the mean difference was 0.12 mm (95% CI 0.06 to 0.18, p <0.0001).



Table 3

Coronary angiographic and intravascular ultrasound (IVUS) guidance












































































































































































































































































































Variable DIPOL AVID Gaster RESIST TULIP OPTICUS SIPS
Lesions (n)
Intravascular ultrasound guidance 83 406 54 79 73 229 166
Nonintravascular ultrasound guidance 80 394 54 76 71 228 190
Stented
Intravascular ultrasound guidance 100% 100% 87% 100% 97% 97% 50%
Nonintravascular ultrasound guidance 100% 100% 85% 100% 96% 99% 50%
Angiography
Lesion length (mm)
Intravascular ultrasound guidance NA 13.0 ± 7.7 13.4 ± 11.7 7.7 ± 3.5 27 ± 9 11.9 ± 5.1 9.7 ± 6.4
Nonintravascular ultrasound guidance NA 13.0 ± 7.8 13.3 ± 9.2 8.0 ± 4.0 29 ± 10 11.6 ± 5.5 9.7 ± 5.5
Reference diameter (mm) §
Intravascular ultrasound guidance 3.21 ± 0.64 3.05 ± 0.64 2.8 ± 0.5 p: .94 ± 0.57; d: 3.06 ± 0.59 2.95 ± 0.57 2.97 ± 0.53 3.01 ± 0.59
Nonintravascular ultrasound guidance 3.19 ± 0.59 3.00 ± 0.54 2.8 ± 0.5 p: 2.88 ± 0.58; d: 2.90 ± 0.49 2.96 ± 0.53 3.01 ± 0.51 3.00 ± 0.7
Preintervention minimum lumen diameter (mm)
Intravascular ultrasound guidance 0.97 ± 0.33 1.11 ± 0.5 1.1 ± 0.6 0.96 ± 0.37 1.02 ± 0.42 0.96 ± 0.35 0.64 ± 0.46
Nonintravascular ultrasound guidance 0.95 ± 0.32 1.09 ± 0.47 1.0 ± 0.5 1.02 ± 0.44 0.99 ± 0.41 0.99 ± 0.34 0.70 ± 0.55
Postintervention minimum lumen diameter (mm)
Intravascular ultrasound guidance 3.34 ± 0.55 2.93 ± 0.55 2.3 ± 0.4 2.48 ± 0.43 3.01 ± 0.40 3.02 ± 0.49 2.49 ± 0.66
Nonintravascular ultrasound guidance 3.06 ± 0.52 2.87 ± 0.48 2.2 ± 0.5 2.46 ± 0.46 2.80 ± 0.31 2.91 ± 0.41 2.38 ± 0.67
Preintervention diameter stenosis (%)
Intravascular ultrasound guidance 69.7 ± 14.2 63.4 ± 14.1 60 ± 17 65 ± 11 65 ± 13 67.6 ± 11.2 79.1 ± 13.8
Nonintravascular ultrasound guidance 70.2 ± 11.4 63.5 ± 14.3 64 ± 18 64 ± 12 65 ± 10 66.7 ± 10.1 76.8 ± 17.6
Postintervention diameter stenosis (%)
Intravascular ultrasound guidance 3.4 ± 2.9 NA 27 ± 10 19 ± 10 12 ± 7 2.8 ± 7.8 18.8 ± 17.3
Nonintravascular ultrasound guidance 8.9 ± 5.4 NA 26 ± 14 19 ± 9 13 ± 9 6.0 ± 8.0 22.5 ± 19.7
Intravascular ultrasound
Achievement of intravascular ultrasound criteria
Intravascular ultrasound guidance NA 44% 64% 80% NA 56% NA
Nonintravascular ultrasound guidance NA 36% 16% 59% NA NA NA
Final minimum stent area (mm 2 )
Intravascular ultrasound guidance 9.8 ± 2.88 7.55 ± 2.82 8.7 ± 2.5 7.95 ± 2.21 NA 8.1 ± 2.3 7.95 ± 2.37
Nonintravascular ultrasound guidance NA 6.9 ± 2.43 6.9 ± 2.1 7.16 ± 2.48 NA NA NA

Data presented as n, %, or mean ± SD.

Reported for target lesion.


Reported for in-stent only, by patient; N = number of patients.


Reported for angiographic follow-up group only in OPTICUS (IVUS n = 229, non-IVUS n = 228).


§ Mean of proximal and distal except for RESIST study, for which proximal (p) and distal (d) were reported separately.




Figure 1


Angiographic results for mean differences in postintervention angiographic minimum lumen diameter (MLD) with 95% CIs for 7 randomized studies. Combined estimates from both random effects (RE) and fixed effects (FE) models shown. IVUS guidance was associated with significantly larger postprocedure angiographic MLD (heterogeneity I 2 = 42.1%, p(Q) = 0.11; p(Effect) <0.001).


Of the 7 studies, 6 had routine angiographic follow-up data available at 6 months and reported the restenosis rates ( Table 4 ). The exception was AVID that did not include routine angiographic follow-up data. As shown in Figure 2 , IVUS guidance was associated with a significantly lower rate of 6-month angiographic restenosis (22% vs 29%, OR 0.64, 95% CI 0.42 to 0.96, p = 0.02).



Table 4

Clinical end points
































































































































































































































End Point DIPOL AVID Gaster RESIST TULIP OPTICUS SIPS Overall
Intention to treat (n)
Intravascular ultrasound 83 375 54 79 73 273 121 1,058
Nonintravascular ultrasound 80 369 54 76 71 275 148 1,073
Revascularization (n) TLR TLR TVR Any TLR PCI + CABG Clinically driven TLR
Intravascular ultrasound 3 (3.6%) 30 (8.1%) 10 (19%) 21 (27%) 7 (10%) 41 (15%) 21 (17%) 133 (13%)
Nonintravascular ultrasound 6 (7.5%) 45 (12%) 18 (33%) 31 (41%) 17 (23%) 38 (14%) 43 (29%) 198 (18%)
Major adverse cardiac events (n) Death, MI, repeat coronary revascularization TLR, death, MI, stent thrombosis, CABG Death, QMI, repeat PCI, CABG Cardiac events Death, MI, TLR Death, QMI, NQMI, repeat PCI, CABG Death, QMI, clinically driven TLR, nonclinically driven revascularization
Intravascular ultrasound 6 (7.3%) 68 (18%) 11 (22%) 20 (25%) 9 (12%) 49 (18%) 37 (30%) 201 (19%)
Nonintravascular ultrasound 13 (16%) 70 (19%) 22 (41%) 28 (37%) 19 (27%) 42 (15%) 55 (37%) 249 (23%)
Mortality (n)
Intravascular ultrasound 1 (1.2%) 12 (3.3%) 0 (0%) 1 (1.3%) 2 (2.7%) 5 (1.8%) 4 (3.3%) 25 (2.4%)
Nonintravascular ultrasound 1 (1.25%) 7 (1.9%) 2 (3.7%) 1 (1.3%) 1 (1.4%) 1 (0.4%) 4 (2.7%) 17 (1.6%)
Myocardial infarction (n) MI MI QMI MI (6 mo) MI MI
Intravascular ultrasound 1 (1.2%) 25 (6.8%) 1 (2%) NA 1 (1%) 6 (2.2%) 1 (0.8%) 35 (3.6%)
Nonintravascular ultrasound 4 (5%) 19 (5.1%) 0 (0%) NA 5 (7%) 10 (3.6%) 6 (3.4%) 44 (4.4%)
Angiographic follow-up (n)
Intravascular ultrasound 73 NA 51 71 64 229 93 581
Nonintravascular ultrasound 67 NA 52 73 61 228 117 598
Restenosis (n)
Intravascular ultrasound 7/73 (10%) NA 8/51 (16%) 16/71 (23%) 15/64 (23%) 56/229 (25%) 27/93 (29%) 129 (22%)
Nonintravascular ultrasound 18/67 (27%) NA 13/52 (25%) 21/73 (29%) 28/61 (46%) 52/228 (23%) 41/117 (35%) 173 (29%)

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

Stay updated, free articles. Join our Telegram channel

Dec 22, 2016 | Posted by in CARDIOLOGY | Comments Off on Meta-Analysis of Randomized Studies Comparing Intravascular Ultrasound Versus Angiographic Guidance of Percutaneous Coronary Intervention in Pre–Drug-Eluting Stent Era

Full access? Get Clinical Tree

Get Clinical Tree app for offline access