Page 92 - Read Online

Page 8 of 18 Browne et al. Vessel Plus 2024;8:19 https://dx.doi.org/10.20517/2574-1209.2023.126

Table 4. Summary of studies examining rates of IMA graft occlusion by harvesting technique*
Patients - n/N (%) Grafts - n/N (%) Study Time of postoperative CABG surgery
Study Conduit
SKT PED SKT PED design angiography type
Calafiore et al., 4/133 1 /71 (1.4) 4/281 1/149 (0.7) BIMA P, SC ≤ 30 days or later (SKT 14.3 On/off-pump
[49]
1999 (3.0) (1.4) mo, PED 7.6 mo)
Amano et al., 1/96 (1.0) 0/76 (0.0) 1/159 0/106 IMA P, MC < 3 months On/off-pump
[48]
2002 (0.6) (0.0)
Hirose et al., 0/87 0/36 (0.0) 0/195 0/77 (0.0) IMA P, MC In-hospital On/off-pump
2003 [50] (0.0) (0.0)
Kai et al., 4/137 0/23 (0.0) 4/274 0/46 (0.0) BIMA RE, SC, SS < 30 days On-pump PED or
2007 [51] (2.9) (1.5) off-pump SKT
† † † †
Mannacio et al., 0/100 0/100 0/100 0/100 LIMA P, R, SC 2 years Off-pump
2011 [45] (0.0) (0.0) (0.0) (0.0)
Sun et al., 9/778 12/795 9/778 12/795 LIMA P, SC 1 year Off-pump
2015 [44] (1.2) (1.5) (1.2) (1.5)
Lamy et al., 28/282 29/720 33/344 30/764 IMA P, MC 1 year On/off-pump
[9]
2021 (9.9) (4.0) (9.6) (3.9)
Dreifaldt et al., 5/52 (9.6) 2/48 (4.2) 5/52 (9.6) 2/48 (4.2) LIMA P, R, SC 3 years On-pump
[54]
2021
Total: 51/1665 44/1869 56/2183 45/2085
(3.1) (2.4) (2.6) (2.2)
95%CI: 2.3-4.0 1.7-3.2 1.9-3.3 1.6-2.9

*Graft occlusion is defined as 100% stenosis. Grafts with “string signs” were not included. Only patients with angiographic results were included.
†
5 patients were excluded after undergoing angiography but from which group was not reported. BIMA: Bilateral internal mammary artery
grafting; LIMA: left internal mammary artery; PED: pedicled; P: prospective; RA: radial artery; R: randomized; RE: retrospective; SC: single-centre;
SKT: skeletonized; SS: single surgeon; MC: multi-centre; NR: not reported.

Subsequently, a large single-center observational study reported comparable rates of graft failure in both
[44]
skeletonized (1.2% [9/778]) and pedicled (1.5% [12/795]) groups among patients who underwent off-pump
CABG and completed the 1-year follow-up angiography. However, the angiographic follow-up period was
notably shorter for the skeletonized LIMA group, averaging 19.4 months compared with 40.0 months for
the pedicled group (P < 0.001), limiting the internal validity of the results.

We recently conducted a post hoc analysis of the COMPASS trial dataset to evaluate the impact of
skeletonized or pedicled IMA harvesting on graft patency and clinical outcomes. The primary outcome was
[9]
graft occlusion, determined by computed tomography angiography . The occlusion rate was higher in
skeletonized IMA compared with pedicled IMA (9.6% [33/344] vs. 3.9% [30/764]; odds ratio: 2.41;
95%CI: 1.39-4.20) 1 year after CABG surgery. This included the left internal mammary artery to the left
anterior descending artery (7.3% [21/289] vs. 3.4% [25/725]; odds ratio: 2.10; 95%CI: 1.14-3.88). The results
were consistent in both LIMA (adjusted odds ratio: 2.13; 95%CI: 1.16-3.91) and RIMA (adjusted odds ratio:
2.88; 95%CI: 0.62-13.49), although relatively few RIMA grafts were evaluated. Regarding clinical events, we
observed that skeletonized harvesting was associated with a higher rate of major adverse cardiovascular
events (hazard ratio, 3.19; 95%CI: 1.53-6.67), mainly driven by an increased risk of repeated
revascularization (hazard ratio, 2.75; 95%CI: 1.10-6.88) after a mean follow-up of 23 months. We did not
observe any mortality benefit from skeletonization (0.4% vs. 0%). Overall, our study suggested that the use
of skeletonized IMA harvesting was associated with a higher rate of graft occlusion and complications
compared with the traditional pedicled technique.

Subsequently, a similar post hoc analysis of the Arterial Revascularization Trial (ART) dataset was
undertaken to assess the impact of skeletonized vs. pedicled IMA on long-term (10 year) clinical
outcomes . The risk of the composite of all-cause mortality, myocardial infarction, or repeat
[8]