Concomitant Versus Staged Percutaneous Cervicocerebral Artery Revascularization with Coronary Artery Bypass Grafting

Abstract

Introduction

The optimal strategy and timing for percutaneous cervicocerebral artery revascularization (PCAR) combined with coronary artery bypass grafting (CABG) remain uncertain in patients with severe carotid or vertebral artery stenosis.

Methods

We retrospectively analyzed adults undergoing concomitant or staged PCAR–CABG at Beijing Anzhen Hospital from January 2011 to December 2025. The primary endpoint was perioperative death or stroke. Strategies were compared using Firth penalized logistic regression. In the staged cohort, restricted cubic splines and interval-based analyses assessed the association between interprocedural interval and outcomes.

Results

Among 801 patients, 57 (7.1%) underwent concomitant and 744 (92.9%) underwent staged intervention; the median interprocedural interval was 30 days (interquartile range, 9–46). PCAR was performed for severe vertebral stenosis in 221 (27.6%), severe carotid stenosis in 474 (59.2%), and both in 106 (13.2%). Perioperative death or stroke occurred in one patient (1.8%) in the concomitant group and 37 patients (5.0%) in the staged group. Concomitant intervention was not significantly associated with the primary endpoint after adjustment (adjusted odds ratio (OR) 0.43; 95% confidence interval (CI), 0.05–1.76; P = 0.280). Exploratory analyses suggested decreasing risk with longer intervals, although no significant nonlinear association was identified. The 1–7-day staged group had the highest estimated odds relative to the concomitant group (adjusted OR, 4.89; 95% CI, 1.04–48.30; P = 0.044).

Conclusions

Perioperative death or stroke did not differ significantly between concomitant and staged PCAR–CABG. However, the limited number of events precludes conclusions regarding equivalence. Interval-specific findings were exploratory and require confirmation in adequately powered prospective multicenter studies.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s40119-026-00465-5.

Key Summary Points

Why carry out this study?

The optimal strategy and timing of percutaneous cervicocerebral artery revascularization (PCAR) combined with coronary artery bypass grafting (CABG) remain uncertain in patients with severe carotid or vertebral artery stenosis.

What was learned from this study?

In this single-center cohort of 801 patients undergoing concomitant or staged PCAR–CABG, perioperative death or stroke occurred in 4.7% of patients, with no statistically significant difference between intervention strategies after adjustment.

Among patients undergoing staged intervention, exploratory analyses suggested a trend toward lower perioperative risk with longer intervals between PCAR and CABG. The 1–7-day subgroup had the highest adjusted odds of perioperative death or stroke relative to concomitant intervention; however, this finding should be interpreted cautiously because of sparse events and potential residual confounding.

Introduction

Polyvascular disease, defined as atherosclerotic involvement of multiple arterial territories, has been increasingly recognized as a marker of advanced systemic atherosclerosis and heightened ischemic risk [1, 2]. In patients with coronary artery disease, atherosclerosis frequently coexists in non-coronary vascular beds, manifesting clinically as cerebrovascular disease, such as prior ischemic stroke or transient ischemic attack, carotid or vertebral artery stenosis, and peripheral arterial disease [3]. This multisite atherosclerotic burden is particularly relevant in patients undergoing coronary artery bypass grafting (CABG). Among patients undergoing CABG, concomitant carotid artery stenosis, especially when severe, bilateral, or symptomatic, has been consistently associated with an increased risk of perioperative stroke compared with patients without carotid disease [4, 5].

In patients with severe carotid artery stenosis requiring CABG, combined carotid and coronary revascularization strategies have been used for more than five decades [6]. Over time, management strategies have evolved from concomitant or staged carotid endarterectomy (CEA) with CABG to hybrid approaches incorporating percutaneous carotid artery revascularization performed before or during CABG [7]. In contrast, evidence guiding the management of severe vertebral artery stenosis in the setting of CABG remains limited. Nevertheless, severe carotid or vertebral artery stenosis may compromise cerebral perfusion, increase susceptibility to perioperative hemodynamic instability, and reflect an increased risk of cerebrovascular events. Accordingly, in selected patients with severe cervicocerebral artery stenosis, including carotid or vertebral artery involvement, concomitant or staged percutaneous cervicocerebral artery revascularization (PCAR) combined with CABG has been performed at our institution.

Despite the clinical relevance of this strategy, several important questions remain unresolved. First, the optimal revascularization strategy for patients requiring both PCAR and CABG remains uncertain, particularly regarding the choice between concomitant and staged procedures. Second, among patients undergoing staged intervention, it remains unclear whether the interval between PCAR and CABG affects the risk of perioperative death or stroke. Accordingly, this study aimed to compare the risk of perioperative death or stroke between concomitant and staged PCAR combined with CABG and to further assess the association between interprocedural interval and clinical outcomes among patients undergoing staged procedures.

Methods

Study Design and Patient Selection

This study involving human participants and human data was approved by the Institutional Review Board of Beijing Anzhen Hospital, Capital Medical University (IRB No. 2026125X), which served as the ethics committee for this single-center study. Given the retrospective design of the study and the use of anonymized clinical data, the requirement for written informed consent was waived by the Institutional Review Board. The study was performed in accordance with the Declaration of Helsinki of 1964 and its later amendments. Between January 1, 2011, and December 31, 2025, approximately 73,000 adult patients underwent CABG at our institution. From this source population, patients aged 18 years or older who underwent percutaneous carotid and/or vertebral artery revascularization (PCAR) in conjunction with CABG, either as a concomitant or staged procedure, were retrospectively identified. During the same period, 67 patients underwent carotid endarterectomy in conjunction with CABG and were not included in the present cohort. Patients were excluded if they had undergone previous cardiac surgery or required additional concomitant cardiac procedures, including valve repair or replacement or ventricular reconstruction. Operative priority was not an exclusion criterion; however, because it was not systematically recorded as elective, urgent, or emergency, the distribution of operative urgency could not be reliably determined. Relevant demographic, clinical, procedural, and outcome data were extracted from the institutional electronic medical record system. The primary endpoint was a composite of perioperative death or stroke. Stroke was defined as a new focal or global neurologic deficit lasting more than 24 hours and confirmed by brain computed tomography or magnetic resonance imaging. Neuroimaging was performed when stroke was clinically suspected on the basis of new postoperative neurological signs or symptoms and was not routinely undertaken in asymptomatic patients.

PCAR and CABG

As part of the routine preoperative evaluation, all CABG candidates at our institution underwent duplex ultrasonographic screening for carotid and vertebral artery stenosis. Patients with suspected stenosis were further evaluated by cervicocerebral computed tomography angiography. Multidisciplinary assessment was performed for patients with relevant neurologic symptoms or a history of transient ischemic attack or stroke, as well as for asymptomatic patients with severe carotid or vertebral artery stenosis, defined as stenosis of 70% or greater. The indication for carotid or vertebral artery revascularization was established through multidisciplinary evaluation led by neurologists, with consideration of neurologic symptoms, stenosis severity, vascular anatomy, and overall procedural risk. PCAR was the predominant revascularization strategy at our institution and was generally selected when revascularization was indicated and an endovascular approach was considered appropriate based on individualized clinical judgment, vascular anatomy, procedural risk, and treating physician expertise. Among patients with carotid artery stenosis, however, no prespecified institutional algorithm governed the choice between PCAR and carotid endarterectomy. Although most patients underwent PCAR, carotid endarterectomy remained an alternative approach and was performed selectively according to the treating surgeon’s expertise and procedural preference. For patients undergoing staged intervention, the interprocedural interval was not protocolized but was determined pragmatically according to clinical considerations, procedural scheduling, and patient availability.

Statistical Analysis

Continuous variables are presented as mean with standard deviation or median with interquartile range, as appropriate, whereas categorical variables are reported as frequencies and percentages. Between-group comparisons were performed using the independent-samples t test or Mann–Whitney U test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables, as appropriate.

Because of the limited number of primary endpoint events and the potential for separation, Firth penalized logistic regression was used. Univariable models were first fitted to assess the association of procedural strategy, defined as concomitant versus staged PCAR–CABG, and baseline characteristics with the primary endpoint. Baseline variables associated with the primary endpoint at P < 0.10 in the univariable analyses were subsequently included as covariates in the multivariable Firth penalized logistic regression model evaluating the association between procedural strategy and the primary endpoint.

Among patients undergoing staged intervention, the association between the interval from PCAR to CABG and the primary endpoint was further explored. Restricted cubic spline analysis was used to evaluate the potential nonlinear association between the interprocedural interval and the primary endpoint. Nonlinearity was evaluated using likelihood-ratio tests. In an additional exploratory analysis, staged procedures were categorized according to interprocedural interval as 1–7, 8–30, 31–90, or > 90 days. These four interval groups were compared with the concomitant group, with concomitant intervention serving as the reference category.

Associations are reported as odds ratios (ORs) with corresponding 95% confidence intervals (CIs). All tests were two-sided, and P value < 0.05 was considered statistically significant. Statistical analyses were performed using R software, version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Characteristics of the Study Population

A total of 801 patients were included in this study. Of these, 57 patients (7.1%) underwent concomitant intervention, whereas 744 patients (92.9%) underwent staged intervention. In the staged cohort, the median interval between PCAR and CABG was 30 days (interquartile range, 9–46 days) (Table 1). The mean age of the overall study population was 65.5 ± 7.0 years, and 132 patients (16.5%) were women. Hypertension was present in 564 patients (70.4%), diabetes mellitus in 374 (46.7%), hyperlipidemia in 512 (63.9%), peripheral artery disease in 117 (14.6%), and prior stroke in 153 (19.1%). PCAR was performed for severe vertebral artery stenosis in 221 patients (27.6%), severe carotid artery stenosis in 474 patients (59.2%), and concomitant severe vertebral and carotid artery stenosis in 106 patients (13.2%). Fifteen of the 801 patients (1.9%) had non-ST-segment elevation myocardial infarction (NSTEMI) as the principal diagnosis at admission, whereas none had ST-segment elevation myocardial infarction. Overall, the composite outcome of perioperative death or stroke occurred in 38 patients (4.7%). Death occurred in seven patients (0.9%), and stroke occurred in 32 patients (4.0%). Peripheral artery disease, chronic renal failure, and CABG performed with cardiopulmonary bypass were associated with a higher risk of perioperative death or stroke (Table 2).

Table 1.

Baseline characteristics of the overall cohort

	Overall cohort (n = 801)
Baseline characteristics
Age, years	65.5 ± 7.0
Female sex (%)	132 (16.5)
Hypertension (%)	564 (70.4)
Diabetes (%)	374 (46.7)
Hyperlipemia (%)	512 (63.9)
Peripheral arterial disease (%)	117 (14.6)
COPD (%)	20 (2.5)
Chronic renal dysfunction (%)	20 (2.5)
Atrial fibrillation (%)	28 (3.5)
Previous stroke (%)	153 (19.1)
Previous PCI (%)	104 (13.0)
NSTEMI on presentation (%)	15 (1.9)
Cerebrovascular intervention type
Vertebral only (%)	221 (27.6)
ICA only (%)	474 (59.2)
Vertebral + ICA (%)	106 (13.2)
Concomitant intervention (%)	57 (7.1)
Staged intervention interval, days	30 [9, 46]
With cardiopulmonary bypass (%)	19 (2.4)

COPD chronic obstructive pulmonary disease, PCI percutaneous coronary intervention, NSTEMI non-ST-segment elevation myocardial infarction, ICA internal carotid artery

Table 2.

Univariate associations of baseline characteristics with perioperative outcomes

Baseline characteristics	OR (95% CI)	P value
Age, years	1.033 (0.984–1.087)	0.195
Female sex (%)	1.211 (0.496–2.612)	0.652
Hypertension (%)	0.786 (0.405–1.591)	0.491
Diabetes (%)	1.280 (0.671–2.458)	0.452
Hyperlipemia (%)	1.802 (0.886–4.005)	0.107
Peripheral arterial disease (%)	2.250 (1.032–4.567)	0.042
COPD (%)	1.527 (0.166–6.260)	0.643
Chronic renal dysfunction (%)	5.909 (1.750–16.501)	0.007
Atrial fibrillation (%)	2.855 (0.740–8.187)	0.116
Previous stroke (%)	1.381 (0.617–2.831)	0.413
Previous PCI (%)	1.628 (0.664–3.530)	0.268
NSTEMI on presentation (%)	2.068 (0.223–8.748)	0.448
Cerebrovascular intervention type	0.898 (0.530–1.511)	0.687
Vertebral only (%)
ICA only (%)
Vertebral + ICA (%)
With cardiopulmonary bypass (%)	11.119 (3.855–29.370)	< 0.001

OR odds ratio, CI confidence interval, COPD chronic obstructive pulmonary disease, PCI percutaneous coronary intervention, NSTEMI non-ST-segment elevation myocardial infarction, ICA internal carotid artery

Concomitant Versus Staged Intervention

Baseline characteristics were generally comparable between the concomitant and staged intervention groups, with no statistically significant between-group differences (Table 3). The composite endpoint of perioperative death or stroke occurred in 1 patient (1.8%) in the concomitant group and 37 patients (5.0%) in the staged group. No perioperative deaths occurred in the concomitant group, whereas one patient experienced a stroke. In the staged group, seven deaths and 31 strokes were observed. After multivariable adjustment, concomitant intervention was not significantly associated with the risk of perioperative death or stroke compared with staged intervention (adjusted OR, 0.43; 95% CI, 0.05–1.76; P = 0.280) (Table 4).

Table 3.

Baseline characteristics of the study population by intervention strategy

Baseline characteristics	Concomitant intervention (n = 57)	Staged intervention (n = 744)	P value
Age, years	64.88 ± 7.00	65.55 ± 6.95	0.482
Female sex (%)	8 (14.0)	124 (16.7)	0.741
Hypertension (%)	38 (66.7)	526 (70.7)	0.623
Diabetes (%)	23 (40.4)	351 (47.2)	0.391
Hyperlipemia (%)	36 (63.2)	476 (64.0)	> 0.999
Peripheral arterial disease (%)	11 (19.3)	106 (14.2)	0.398
COPD (%)	0 (0.0)	20 (2.7)	0.389
Chronic renal dysfunction (%)	0 (0.0)	20 (2.7)	0.389
Atrial fibrillation (%)	4 (7.0)	24 (3.2)	0.131
Previous stroke (%)	13 (22.8)	140 (18.8)	0.573
Previous PCI (%)	6 (10.5)	98 (13.2)	0.713
NSTEMI on presentation (%)	0 (0.0)	15 (2.0)	0.617
Cerebrovascular intervention type			0.343
Vertebral only (%)	16 (28.1)	205 (27.6)
ICA only (%)	37 (64.9)	437 (58.7)
Vertebral + ICA (%)	4 (7.0)	102 (13.7)
With cardiopulmonary bypass (%)	3 (5.3)	16 (2.2)	0.147

COPD chronic obstructive pulmonary disease, PCI percutaneous coronary intervention, NSTEMI non-ST-segment elevation myocardial infarction, ICA internal carotid artery

Table 4.

Unadjusted and adjusted effects of intervention strategy (concomitant vs. staged) on perioperative outcomes

	Concomitant intervention (n = 57)	Staged intervention (n = 744)	Unadjusted OR		Adjusted OR
			OR (95% CI)	P value	OR (95% CI)	P value
Death or stroke	1 (1.8)	37 (5.0)	0.50 (0.06–1.93)	0.364	0.43 (0.05–1.76)	0.280
Death	0 (0.0)	7 (0.9)	0.86 (0.01–7.17)	0.913	0.54 (0.00–5.13)	0.664
Stroke	1 (1.8)	31 (4.2)	0.60 (0.07–2.34)	0.518	0.57 (0.06–2.30)	0.484

OR odds ratio, CI confidence interval. Data are presented as n (%). Odds ratios compare concomitant intervention with staged intervention, with staged intervention as the reference group. The adjusted model included chronic renal dysfunction, peripheral artery disease, and cardiopulmonary bypass. Firth penalized logistic regression was used because of the sparse endpoint events and the potential for small-sample bias and separation

Impact of Interprocedural Interval in the Staged Cohort

Among patients undergoing staged intervention, restricted cubic spline analysis did not identify a statistically significant nonlinear association between the interprocedural interval and the risk of perioperative death or stroke. Nevertheless, the spline curve suggested a trend toward lower risk with increasing intervals between PCAR and CABG (Fig. 1). No statistically significant differences in baseline characteristics were observed among the concomitant, 1–7-day, 8–30-day, 31–90-day, and > 90-day groups (Supplementary Table 1). The primary endpoint occurred in 1 of 57 patients (1.8%) in the concomitant group, 10 of 120 (8.3%) in the 1–7-day group, 12 of 228 (5.3%) in the 8–30-day group, 13 of 344 (3.8%) in the 31–90-day group, and 2 of 52 (3.8%) in the > 90-day group. With the concomitant group as the reference, adjusted odds ratio point estimates for perioperative death or stroke generally decreased across staged-intervention groups with progressively longer interprocedural intervals (Table 5). The 1–7-day group had the highest adjusted odds relative to the concomitant group (adjusted OR, 4.89; 95% CI, 1.04–48.30; P = 0.044). In an additional analysis further adjusting for NSTEMI at presentation, the 1–7-day group continued to have the highest adjusted point estimate (adjusted OR, 4.79; 95% CI, 1.01–47.38; P = 0.049) (Supplementary Table 2).

Fig. 1.

Association between the PCAR–CABG interval and the odds of perioperative death or stroke modeled using restricted cubic splines. Association between the interval from percutaneous cervicocerebral artery revascularization to coronary artery bypass grafting and perioperative death or stroke. The association was modeled using restricted cubic spline analysis. The solid red line represents the estimated odds ratio (OR), and the shaded area represents the 95% confidence interval (CI). The dashed horizontal line indicates an OR of 1

Table 5.

Adjusted associations between intervention timing strategy (concomitant vs. staged at different intervals) and perioperative outcomes

	Concomitant intervention (n = 57)	Staged intervention
		1–7 days (n = 120)		8–30 days (n = 228)		31–90 days (n = 344)		> 90 days (n = 52)
	n (%)	n (%)	Adjusted OR (95% CI)	n (%)	Adjusted OR (95% CI)	n (%)	Adjusted OR (95% CI)	n (%)	Adjusted OR (95% CI)
Death or stroke	1 (1.8)	10 (8.3)	4.89 (1.04–48.30)	12 (5.3)	2.64 (0.58–25.69)	13 (3.8)	1.75 (0.39–16.80)	2 (3.8)	1.80 (0.22–20.89)
Death	0 (0.0)	5 (4.2)		0 (0.0)		1 (0.3)		1 (1.9)
Stroke	1 (1.8)	6 (5.0)		12 (5.3)		12 (3.5)		1 (1.9)

OR odds ratio, CI confidence interval. Data are presented as n (%). Odds ratios compare each staged interprocedural-interval subgroup with concomitant intervention, with concomitant intervention as the reference group. The model included four indicator terms representing the staged interval categories and was adjusted for peripheral arterial disease, chronic renal dysfunction, and cardiopulmonary bypass. Firth penalized logistic regression was used because of the sparse endpoint events and the potential for small-sample bias and separation. Given the limited number of events relative to the number of model coefficients, these estimates should be considered exploratory

Discussion

The present study has two principal findings. First, no statistically significant difference in the risk of perioperative death or stroke was observed between concomitant and staged PCAR–CABG strategies. Second, among patients undergoing staged intervention, exploratory analyses suggested a trend toward lower perioperative risk with longer intervals between PCAR and CABG, with the 1–7-day subgroup showing the highest estimated odds of perioperative death or stroke relative to the concomitant group.

These findings should, however, be interpreted with caution. The absence of a statistically significant difference between concomitant and staged strategies should not be interpreted as evidence of comparable safety. The limited number of primary endpoint events substantially reduced statistical power and precluded any conclusion regarding equivalence between the two strategies. Moreover, perioperative myocardial infarction was not systematically assessed in the staged PCAR–CABG group and therefore could not be incorporated into the overall comparison of perioperative safety. The interval-specific findings should likewise be regarded as exploratory and hypothesis-generating, given the small number of events in each interval subgroup and the wide confidence intervals, which indicate considerable statistical uncertainty.

CABG remains the preferred revascularization strategy for patients with complex coronary artery disease, left ventricular dysfunction, or diabetes mellitus [8]. Although perioperative stroke occurs in only approximately 1–2% of patients undergoing CABG [9–11], it remains one of the most serious postoperative complications and is associated with substantially increased perioperative mortality, morbidity, and healthcare costs [12, 13]. The pathogenesis of perioperative stroke after CABG is multifactorial and may involve embolization of atherosclerotic debris from the ascending aorta during cannulation, aortic cross-clamping, or proximal anastomosis; microembolization related to cardiopulmonary bypass; transient intraoperative hypotension; and postoperative new-onset atrial fibrillation. In patients with carotid artery stenosis, unstable atherosclerotic plaques may serve as a potential source of thromboembolism. In addition, carotid stenosis may compromise cerebral perfusion reserve, thereby increasing susceptibility to cerebral hypoperfusion during episodes of intraoperative hypotension [14]. In this context, current guidelines recommend carotid revascularization in patients undergoing CABG who have symptomatic moderate-to-severe carotid artery stenosis, as well as in selected asymptomatic patients with bilateral severe carotid stenosis or unilateral severe stenosis accompanied by contralateral carotid occlusion [15]. Although evidence regarding the management of severe vertebral artery stenosis in patients undergoing CABG remains limited, the vertebral arteries are major contributors to the posterior cerebral circulation, suggesting that severe vertebral artery disease may also be relevant to perioperative cerebrovascular risk assessment.

In patients with severe carotid artery stenosis requiring CABG, CEA and carotid artery stenting (CAS) are the two principal carotid revascularization strategies. Both approaches can be performed either concomitantly with CABG or as staged procedures. However, concomitant CEA–CABG may increase procedural complexity and operative time, whereas staged CEA–CABG requires two separate procedures under general anesthesia. These limitations have contributed to growing interest in CAS combined with CABG as a less invasive alternative for selected patients. Previous studies have generally found no significant difference in the composite outcome of death, myocardial infarction, or stroke between CAS–CABG and CEA–CABG strategies [16–18]. However, findings from large administrative databases have been inconsistent. Using Nationwide Inpatient Sample (NIS) data from 2000 to 2004, Timaran et al. reported a lower incidence of postoperative stroke with CAS–CABG than with CEA–CABG [19]. In contrast, Feldman et al., analyzing NIS data from 2004 to 2012, reported lower mortality but a higher risk of stroke with CAS–CABG compared with staged or concomitant CEA–CABG [20]. Furthermore, Yang et al. found a similar composite incidence of stroke, myocardial infarction, or death between concomitant and staged CAS–CABG [21].

The present study extends previous evidence by including not only patients with severe carotid artery stenosis requiring CABG but also those with severe vertebral artery stenosis. In this cohort, most patients underwent staged PCAR–CABG, whereas only 7.1% underwent concomitant intervention. Consistent with previous studies, no statistically significant difference in perioperative death or stroke was observed between the concomitant and staged strategies. Importantly, we also explored whether the interval between PCAR and CABG was associated with perioperative outcomes among patients undergoing staged intervention. Adjusted odds ratio point estimates for perioperative death or stroke generally declined across groups with progressively longer interprocedural intervals. In contrast, the 1–7-day subgroup had the highest point estimate relative to the concomitant group.

Several potential mechanisms may account for the higher point estimate observed in the 1–7-day subgroup. First, shortly after PCAR, the treated vascular segment may remain biologically vulnerable because of incomplete endothelialization, procedure-related vascular injury, platelet activation, and residual plaque vulnerability, which may increase susceptibility to thromboembolic events during subsequent CABG. Second, CABG performed soon after PCAR may expose patients to substantial hemodynamic fluctuations, cardiopulmonary bypass-related microembolization, and transient intraoperative hypotension before cerebral perfusion reserve has fully stabilized. Third, antithrombotic management during this early period is particularly challenging. When CABG is performed shortly after PCAR, dual antiplatelet therapy may need to be interrupted or withheld because of concerns about surgical bleeding, whereas bridging with low-molecular-weight heparin may not provide equivalent protection against platelet-mediated stent thrombosis or other thromboembolic events. By contrast, a longer interval between PCAR and CABG may permit greater vascular stabilization, more appropriate management of antithrombotic therapy, and recovery from procedure-related inflammatory and hemodynamic stress. These mechanistic explanations remain speculative. Because the reasons for selecting a 1–7-day interval were not consistently documented, we cannot determine whether earlier CABG was prompted by greater clinical urgency. Patients undergoing CABG within 1–7 days may have had more unstable coronary disease or a higher underlying risk profile. Although additional adjustment for NSTEMI at presentation did not materially alter the association, adjustment for this single covariate could not fully account for differences in clinical urgency or disease severity. Confounding by indication and residual confounding therefore cannot be excluded. Moreover, the limited number of primary endpoint events resulted in wide confidence intervals and substantial imprecision in the effect estimates. Accordingly, the observed association should be regarded as exploratory and hypothesis-generating rather than as evidence that a shorter interprocedural interval causally increases perioperative risk.

Limitations

This study has several limitations. First, perioperative myocardial infarction was not systematically assessed. The risk of myocardial infarction during the interval between PCAR and CABG is an important consideration when comparing concomitant and staged strategies. Second, this was a single-center study, and institutional differences in procedural techniques and perioperative management may limit the generalizability of our findings. Third, the retrospective design is inherently susceptible to selection bias and residual confounding due to unmeasured factors, such as operative urgency. In addition, the timing of staged intervention was not governed by a prespecified protocol but was largely determined by procedural scheduling and patient availability, which may have introduced further bias into the interval-specific analyses. Fourth, postoperative neuroimaging was performed only when stroke was clinically suspected and was not routinely obtained in all patients. Therefore, clinically silent or subtle cerebral infarctions may have gone undetected. In addition, stroke severity was not systematically assessed using validated scales such as the National Institutes of Health Stroke Scale or the modified Rankin Scale, precluding reliable classification of strokes as minor versus major. Finally, the limited number of primary endpoint events reduced statistical power and limited the precision of the effect estimates. Therefore, large, prospective, multicenter studies with comprehensive assessment of perioperative outcomes, including myocardial infarction, are warranted to validate these findings.

Conclusions

In this single-center retrospective cohort study, no statistically significant difference in perioperative death or stroke was observed between concomitant and staged PCAR–CABG strategies. Among patients undergoing staged intervention, exploratory analyses suggested a trend toward lower perioperative risk with longer intervals between PCAR and CABG, whereas the 1–7-day subgroup had the highest estimated odds of perioperative death or stroke relative to the concomitant group. Given the limited number of events and the resulting imprecision of the estimates, these findings should be interpreted cautiously and require validation in large, prospective, multicenter studies.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Jinwei Zhang and Liaoming He contributed equally to this work. They conceived and designed the study, performed the statistical analyses, and jointly drafted the original manuscript. Peng Sun and Chenzhen Xu participated in data collection and preliminary statistical analyses. Chuanjie Yue and Song Wang independently verified the data and analytical results. Hao Cui, Changwei Ren, and Yongqiang Lai supervised the study and were responsible for its conceptualization, design, implementation, project administration, and overall oversight. All authors critically reviewed and revised the manuscript, approved the final version for publication, and agreed to be accountable for all aspects of the work. All listed authors meet the authorship criteria, and no eligible contributors have been omitted.

Funding

This study was jointly supported by the Excellent Youth Fund of the National Natural Science Foundation of China, the Excellent Youth Fund of Capital Medical University (grant No. KCA-2305), and the Beijing Natural Science Foundation (grant No. L251020). The funders had no role in the study design or conduct; data collection, management, analysis, or interpretation; manuscript preparation, review, or approval; or the decision to submit the manuscript for publication. No funding was received for the publication of this study.

Data Availability

All reasonable requests for data should be submitted to the corresponding author for consideration.

Declarations

Conflict of Interests

Jinwei Zhang, Liaoming He, Chenzhen Xu, Chuanjie Yue, Song Wang, Peng Sun, Hao Cui, Changwei Ren and Yongqiang Lai declare no competing interests.

Ethical Approval

This study involving human participants and human data was approved by the Institutional Review Board of Beijing Anzhen Hospital, Capital Medical University (IRB No. 2026125X), which served as the ethics committee for this single-center study. Given the retrospective design of the study and the use of anonymized clinical data, the requirement for written informed consent was waived by the Institutional Review Board. The study was performed in accordance with the Declaration of Helsinki of 1964 and its later amendments.