Abstract
Background
Long-term survival following off-pump coronary artery bypass grafting (OPCAB) compared with conventional on-pump coronary artery bypass grafting (ONCAB) remains uncertain. Randomised trials and meta-analyses up to 5–10 years after procedures have demonstrated either no significant differences or modest disadvantages of OPCAB, particularly regarding completeness of revascularisation and late mortality. However, comparative data extending up to 20 years after procedure are scarce. Thus, very-long-term outcomes after ONCAB versus OPCAB in a propensity score-matched cohort were evaluated.
Methods
Consecutive adults undergoing isolated primary coronary artery bypass grafting between 1999 and 2016 at a single cardiac surgery centre were retrospectively identified. Patients undergoing isolated OPCAB or ONCAB procedures were included and matched in a 1:2 ratio using clinically relevant preoperative variables. The primary outcome was overall survival in the matched cohort analysed using Cox proportional hazards regression.
Results
During the study period, 2,935 patients underwent isolated primary coronary artery bypass grafting (CABG), of whom 2,496 (85.0%) underwent ONCAB and 439 (15.0%) OPCAB procedures. Median survival time was 13.7 years (95% CI 13.2–14.2) and median follow-up was 12.3 years (interquartile range 8.0–17.1, maximum 23.9 years). The matched cohort consisted of 402 OPCAB patients matched to 804 ONCAB patients. Median survival time was 12.4 years (95% CI 11.9–13.2) in ONCAB and 12.6 years (95% CI 11.5–13.8) in OPCAB patients. Overall survival did not differ significantly between the groups (HR 0.95, 95% CI 0.82–1.11; ONCAB as reference). No significant differences were observed in the cumulative incidence of cardiovascular mortality.
Conclusion
In this propensity score-matched cohort with follow-up extending up to 20 years, ONCAB and OPCAB were associated with comparable long-term survival and cardiovascular mortality. These findings support an individualised approach to operative technique selection based on patient characteristics and surgical expertise.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12872-026-06117-9.
Keywords: Coronary artery bypass grafting, Cardiopulmonary bypass, Off-pump surgery, Cardiothoracic surgery, Cardiac surgery
Background
Coronary artery disease accounts for approximately 13% of all deaths and remains the leading cause of death globally [1]. Coronary artery bypass grafting (CABG) is an established surgical treatment for multivessel coronary artery disease. CABG can be performed either with cardiopulmonary bypass (on-pump CABG, ONCAB) or without it (off-pump CABG, OPCAB). The conventional ONCAB technique facilitates easier cardiac manipulation compared with the OPCAB approach. Long-term series report 15–22-year survival rates of approximately 20–33% after CABG, reflecting both the progression of underlying disease and the aging of the surgical population [2].
OPCAB technique was designed to avoid cardiopulmonary bypass (CPB) and cardioplegic arrest, with the aim of reducing systemic inflammatory response, neurologic injury, renal dysfunction and other CPB-related complications [3, 4]. However, OPCAB is technically demanding and may be associated with fewer distal anastomoses and higher rates of incomplete revascularisation, potentially compromising long-term graft patency and survival [5–9].
Contemporary guidelines do not recommend either technique as superior, but in a heavily calcified aorta, OPCAB with the “aortic no-touch” technique is reasonable [10, 11]. Several randomised and observational studies have reported either no difference between the techniques or a modest advantage for ONCAB [4, 6–8, 12–14]. Very-long-term comparative data extending up to 20 years remains limited, particularly in low-volume centres. Large, randomised trials have generally demonstrated no short-term survival benefit of OPCAB and, in some cases, have suggested worse mid-term outcomes [4, 8, 15, 16]. OPCAB has also been associated with inferior outcomes in low-volume settings [17]. Meta-analyses of randomised and observational studies have shown either no difference or slightly higher long-term mortality with OPCAB, especially in lower-volume centres [12–14, 18–20]. In contrast, high-volume OPCAB centres have reported comparable or even favourable long-term survival when complete revascularisation is achieved [7, 17, 21].
Many previous observational studies relied on conventional multivariable adjustment, whereas few have applied propensity score methodologies with extended long-term follow-up. Therefore, outcomes were compared up to 20 years after ONCAB and OPCAB in a single-centre cohort using propensity score matching to balance baseline characteristics. Survival after OPCAB was hypothesised to be comparable with ONCAB.
Methods
A retrospective single-centre study was conducted using propensity score matching to compare outcomes between the treatment groups. Patients were identified from the prospectively maintained institutional cardiac surgery database including all CABG patients between 1999 and 2016. Follow-up data were collected retrospectively. Mortality data were collected from the national registry and were unavailable for 178 patients; thus, these patients were excluded. Complete-case analysis was used because the overall extent of missing data was limited. Concomitant valve, aortic, or other major cardiac procedures (n = 468) and redo surgeries (n = 146) were excluded. The study was approved by the institutional review board with a waiver of individual informed consent due to the retrospective nature of the study. This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [22].
The choice between ONCAB and OPCAB was left to the discretion of the attending surgeon, based on patient risk profile, coronary anatomy and intraoperative considerations. Intraoperative transoesophageal ultrasound and epi-aortic ultrasound were routinely used. ONCAB procedures were performed via median sternotomy with standard institution-specific protocols for CPB, aortic cannulation and cardioplegic arrest. Myocardial protection was achieved using tepid blood cardioplegia. Most commonly, cardiac arrest was achieved via an antegrade potassium-magnesium bolus, followed by continuous retrograde blood cardioplegia via coronary sinus for the remainder of the procedure. After completion of distal anastomoses, grafts were perfused with the same cardioplegic solution.
OPCAB procedures were also performed via median sternotomy using mechanical stabilisers and positioning devices. Intracoronary shunts were used at the surgeon’s discretion. Haemodynamic stability was maintained via volume loading, vasoactive support and temporary pacing as needed. The left internal thoracic artery (LITA) was the preferred conduit for the left anterior descending artery (LAD) in both groups. Additional arterial conduits, including the right internal thoracic artery (RITA), radial artery and saphenous vein grafts were used as clinically indicated.
Perioperative clinical variables including age, sex, body surface area, diabetes mellitus, prior myocardial infarction, renal function, chronic obstructive pulmonary disease, peripheral arterial disease, cerebrovascular disease and EuroSCORE II [23], in addition to 30-day perioperative outcomes, were prospectively recorded in the institutional database.
Analyses were performed using R version 4.4.3 [1] with the MatchIt, cobalt, survival and cmprsk packages. A two‑sided p‑value < 0.05 was considered statistically significant. Comparisons of perioperative characteristics between OPCAB and ONCAB were made using the Wilcoxon rank‑sum test for continuous variables and Pearson’s chi‑squared test or Fisher's exact test for categorical variables. Continuous data are presented as medians with interquartile ranges and categorical data as counts with percentages.
Propensity score matching was used to reduce selection bias and estimate the average treatment effect of different operative techniques on outcomes. Only patients with complete covariate data were included in the matched analysis; thus, 37 OPCAB patients and 373 ONCAB patients were excluded before matching. Nearest neighbour pair matching was used in a 1:2 ratio (OPCAB:ONCAB) without replacement and with no calliper restriction. The propensity score was calculated with a logistic regression model including 14 clinically selected preoperative covariates listed in Fig. 1. Intraoperative and postoperative variables were considered post-treatment mediators rather than baseline confounders and were therefore excluded from matching. Balance was assessed using absolute standardised mean differences and was visualised with a Love plot (Fig. 1, supplementary material). Complete case analysis was used because missing data was limited and considered unlikely to materially affect the findings.
Fig. 1.
Covariate balance before and after propensity score matching, expressed as absolute standardized mean differences. LV EF = Left Ventricle Ejection Fraction, CCS/NYHA grade = functional symptom status according to either the Canadian Cardiovascular Society grading or the New York Heart Association grading, COPD = Chronic Obstructive Pulmonary Disease, CKD = Chronic Kidney disease
The primary outcome was overall survival from the date of surgery. Patients were followed until January 2024. Kaplan–Meier estimates were used to derive survival curves and median survival times, with comparisons between groups performed using the log‑rank test. In addition, univariate Cox proportional hazards models were fitted to obtain hazard ratios (HR) and 95% confidence intervals (CI). Proportional hazards assumption was evaluated using Schoenfeld residuals. All outcome analyses were conducted with the propensity score-matched cohort.
Cardiovascular mortality, with non-cardiovascular death treated as a competing risk, was analysed as a secondary outcome. Cumulative incidence functions were plotted and compared using Gray’s test and cumulative incidence at a fixed 10-year time-point was additionally reported. A Fine–Gray competing risks regression model was used and results are reported as sub-distribution HRs with corresponding CIs. Cardiovascular death was defined as I00–I99 of the International Classification of Diseases, 10th Revision, coded as the primary cause of death on death certificates. Early (30-day) post-operative outcomes including all-cause mortality, reoperation, deep sternal wound infection (DSWI), intra-aortic balloon pump (IABP) use, need for haemodialysis and length of hospital stay (LOS) were analysed as secondary endpoints.
Results
All patients
During the study period, a total of 2,935 patients underwent isolated primary CABG, of whom 2,496 (85.0%) received ONCAB and 439 (15.0%) OPCAB. The annual number of OPCAB procedures ranged from 6 to 62, with a median of 22 per year (supplementary material). Overall median follow-up time was 12.3 years (IQR 8.0–17.1, maximum 24.9 years). During follow-up, 1,953 patients (67%) died, including 258 (59%) in the OPCAB group and 1,695 (68%) in the ONCAB group. Cardiovascular deaths occurred in 896 ONCAB patients (35.9%) and 117 OPCAB patients (26.7%). In the overall cohort, the median survival time was 13.8 years (95% CI 13.4–14.4) for ONCAB and 12.8 years (95% CI 11.9–14.0) for OPCAB.
Matched patients
A total of 402 OPCAB patients were matched to 804 ONCAB patients. Adequate post-matching covariate balance was achieved (Fig. 1, Table 1, supplementary material).
Table 1.
Patient characteristics
| Characteristic | Unmatched | ||||
|---|---|---|---|---|---|
|
ONCAB N = 2,496 |
OPCAB N = 439 |
p-value |
ONCAB N = 804 |
OPCAB N = 402 |
|
| Age, years | 68 (30–89) | 70 (38–90) | 0.026 | 70 (30–87) | 70 (38–90) |
| Sex, female | 620 (25%) | 105 (24%) | 0.7 | 200 (25%) | 99 (25%) |
| Euroscore II | 1.48 (0.95–2.59) | 1.46 (0.94–2.79) | 0.6 | 1.58 (0.98, –2.83) | 1.46 (0.98, –2.81) |
| NYHA | 0.024 | ||||
| 1 | 76 (4.2%) | 23 (6.6%) | 37 (6.2%) | 20 (6.1%) | |
| 2 | 486 (27%) | 112 (32%) | 199 (33%) | 105 (32%) | |
| 3 | 815 (45%) | 135 (38%) | 241 (40%) | 128 (39%) | |
| 4 | 446 (24%) | 81 (23%) | 120 (20%) | 73 (22%) | |
| Missing data | 673 | 88 | 207 | 76 | |
| CCS | < 0.001 | ||||
| 1 | 94 (4.0%) | 26 (6.1%) | 43 (5.6%) | 23 (5.9%) | |
| 2 | 556 (23%) | 137 (32%) | 253 (33%) | 129 (33%) | |
| 3 | 1,014 (43%) | 155 (37%) | 279 (36%) | 144 (37%) | |
| 4 | 704 (30%) | 105 (25%) | 190 (25%) | 96 (24%) | |
| Missing data | 128 | 16 | 39 | 10 | |
| Diabetes | 536 (21%) | 85 (19%) | 0.3 | 164 (20%) | 78 (19%) |
| Cerebrovascular disease | 248 (9.9%) | 43 (9.8%) | > 0.9 | 71 (8.8%) | 39 (9.7%) |
| COPD | 248 (9.9%) | 47 (11%) | 0.7 | 79 (9.8%) | 41 (10%) |
| Extracardiac atherosclerosis | 230 (9.2%) | 57 (13%) | 0.018 | 115 (14%) | 51 (13%) |
| Poor mobility | 29 (1.2%) | 16 (3.6%) | < 0.001 | 25 (3.1%) | 16 (4.0%) |
| Recent MI | 684 (27%) | 113 (26%) | 0.5 | 214 (27%) | 105 (26%) |
| LV function | 0.047 | ||||
| > 50% | 1,608 (68%) | 295 (70%) | 570 (71%) | 284 (71%) | |
| 31–50% | 639 (27%) | 115 (27%) | 217 (27%) | 110 (27%) | |
| 21–30% | 108 (4.5%) | 8 (1.9%) | 11 (1.4%) | 6 (1.5%) | |
| < 21% | 24 (1.0%) | 2 (0.5%) | 6 (0.7%) | 2 (0.5%) | |
| Missing data | 117 | 19 | |||
| Glomerular filtration rate | 80 (66–92) | 80 (65–91) | 0.2 | 78 (64–91) | 80 (65–91) |
| Missing data | 231 | 11 | |||
| Anaemia | 459 (18%) | 95 (22%) | 0.13 | 176 (22%) | 88 (22%) |
| Missing data | 9 | 0 | |||
Age is reported as mean (minimum–maximum), and all other continuous variables as median (Q1–Q3). Categorical variables are summarized as n (%). Between‑group comparisons were conducted using the Wilcoxon rank‑sum test for continuous variables and either Pearson’s chi‑square test or Fisher’s exact test for categorical variables, where appropriate. Comparisons are not applicable for matched patients with matched variables
ONCAB On-pump coronary artery bypass grafting, OPCAB Off-pump coronary artery bypass grafting, NYHA The New York Heart association classification, CCS The Canadien Cardiovascular Society classification, COPD Chronic obstructive pulmonary disease, Recent MI Myocardial infarction during the previous 90 days, LV function Left ventricle ejection fraction
The median number of distal anastomoses was four in the ONCAB group and three in OPCAB group (p < 0.001). The ONCAB group also had more distal venous and arterial anastomoses. Internal thoracic artery (ITA) usage did not statistically differ between the two groups (p = 0.093). Among ONCAB patients, median CPB time was 95 min and aortic cross clamp (ACC) time 74 min (Table 2). Thirty-day mortality was low and did not differ significantly between ONCAB and OPCAB patients (1.7% vs 0.5%, p = 0.5). Rates of perioperative myocardial infarction, stroke, reoperation for bleeding, postoperative atrial fibrillation (POAF), IABP, DSWI, pneumonia and renal replacement therapy did not statistically significantly differ between groups. Peak blood creatinine and troponin T (TnT) levels were statistically higher in the ONCAB group (Table 3).
Table 2.
Intraoperative factors
| Characteristic | Unmatched | Matched | ||||
|---|---|---|---|---|---|---|
|
ONCAB N = 2,496 |
OPCAB N = 439 |
p-value |
ONCAB N = 804 |
OPCAB N = 402 |
p-value | |
| Urgency | 0.7 | 0.7 | ||||
| Elective | 1,819 (73%) | 321 (73%) | 600 (75%) | 294 (73%) | ||
| Emergent | 17 (0.7%) | 2 (0.5%) | 0 (0%) | 0 (0%) | ||
| Urgent | 638 (26%) | 110 (25%) | 191 (24%) | 103 (26%) | ||
| Salvage | 22 (0.9%) | 6 (1.4%) | 13 (1.6%) | 5 (1.2%) | ||
| CPB time | 96 (80–113) | NA | NA | 95 (79–112) | NA | NA |
| Missing data | 3 | NA | 2 | NA | ||
| ACC time | 75 (60–90) | NA | NA | 75 (59–89) | NA | NA |
| Missing data | 3 | NA | 3 | NA | ||
| Distal anastomoses | 4 (3–5) | 3 (2–3) | < 0.001 | 4 (3–5) | 3 (2–3) | < 0.001 |
| Missing data | 4 | 1 | ||||
| Venous distal anastomoses | 3 (2–4) | 1 (1–2) | < 0.001 | 3 (2–4) | 1 (1–2) | < 0.001 |
| Missing data | 7 | 3 | 3 | 1 | ||
| Arterial grafts | 1 (1–2) | 1 (1–1) | < 0.001 | 1 (1–1) | 1 (1–1) | < 0.001 |
| Missing data | 5 | 4 | ||||
| ITA usage | < 0.001 | 0.040 | ||||
| No | 162 (6.5%) | 35 (8.2%) | 69 (8.6%) | 43 (11%) | ||
| LITA | 1,937 (78%) | 359 (84%) | 655 (81%) | 334 (83%) | ||
| RITA | 29 (1.2%) | 4 (0.9%) | 3 (0.4%) | 3 (0.7%) | ||
| BITA | 354 (14%) | 27 (6.4%) | 77 (9.6%) | 22 (5.5%) | ||
| Missing data | 14 | 14 | 0 | 0 | ||
| Intraoperative allogenic blood products | < 0.001 | < 0.001 | ||||
| 0 | 1,960 (79%) | 395 (90%) | 635 (79%) | 361 (90%) | ||
| 1–2 | 405 (16%) | 31 (7.1%) | 131 (16%) | 29 (7.2%) | ||
| 3–4 | 99 (4.0%) | 9 (2.1%) | 27 (3.4%) | 8 (2.0%) | ||
| > 4 | 32 (1.3%) | 4 (0.9%) | 11 (1.4%) | 4 (1.0%) | ||
| Intraoperative bleeding | 350 (300–500) | 300 (200–400) | < 0.001 | 300 (280–400) | 300 (200–400) | < 0.001 |
| Missing data | 3 | 0 | 2 | 0 | ||
Continuous variables are reported as median (Q1–Q3). Categorical variables are summarized as n (%). Between‑group comparisons were conducted using the Wilcoxon rank‑sum test for continuous variables and either Pearson’s chi‑square test or Fisher’s exact test for categorical variables, where appropriate
ONCAB On-pump coronary artery bypass grafting, OPCAB Off-pump coronary artery bypass grafting, CPB cardio-pulmonary bypass, ACC aortic cross clamp, ITA internal thoracic artery, LITA left internal thoracic artery, RITA right internal thoracic artery, BITA bilateral thoracic artery
Table 3.
Post-operative outcomes
| Characteristic | Unmatched | Matched | ||||
|---|---|---|---|---|---|---|
|
ONCAB N = 2,496 |
OPCAB N = 439 |
p-value |
ONCAB N = 804 |
OPCAB N = 402 |
p-value | |
| Bleeding from drains | 500 (360–690) | 600 (450–850) | < 0.001 | 520 (370–690) | 598 (450–810) | < 0.001 |
| Missing data | 4 | 0 | 3 | 0 | ||
| Haemoglobin at discharge (g/l) | 102 (94–112) | 108 (98–117) | < 0.001 | 103 (95–113) | 108 (98–117) | < 0.001 |
| Missing data | 27 | 1 | 8 | 0 | ||
| Reoperation (any cause) | 117 (4.7%) | 24 (5.5%) | 0.6 | 42 (5.2%) | 21 (5.2%) | > 0.9 |
| Missing data | 2 | 0 | 2 | 0 | ||
| Reoperation (due to bleeding) | 69 (2.8%) | 14 (3.2%) | 0.7 | 26 (3.2%) | 11 (2.7%) | 0.8 |
| Missing data | 2 | 0 | 2 | 0 | ||
| LOS (days) | 6.00 (5.00–8.00) | 7.00 (5.00–8.00) | 0.5 | 7.00 (5.00–8.00) | 7.00 (5.00–8.00) | > 0.9 |
| Missing data | 19 | 1 | 8 | 1 | ||
| ICU time (days) | 1.00 (1.00–1.00) | 1.00 (1.00–1.00) | 0.6 | 1.00 (1.00–1.00) | 1.00 (1.00–1.00) | 0.7 |
| Missing data | 4 | 0 | 3 | 0 | ||
| Extubation time (hours) | 5.50 (4.50–7.00) | 5.00 (4.00–6.50) | < 0.001 | 5.50 (4.50–7.00) | 5.00 (4.00–6.50) | 0.001 |
| Missing data | 15 | 1 | 6 | 1 | ||
| 30-day mortality | 35 (1.4%) | 2 (0.5%) | 0.2 | 14 (1.7%) | 2 (0.5%) | 0.13 |
| Cerebral infarction | 57 (2.3%) | 8 (1.8%) | 0.7 | 18 (2.3%) | 6 (1.5%) | 0.5 |
| Missing data | 4 | 0 | 4 | 0 | ||
| TIA | 19 (0.8%) | 0 (0%) | 0.10 | 5 (0.6%) | 0 (0%) | 0.2 |
| Missing data | 4 | 0 | 4 | 0 | ||
| AMI | 39 (1.6%) | 8 (1.8%) | 0.9 | 15 (1.9%) | 6 (1.5%) | 0.8 |
| Missing data | 7 | 0 | 3 | 0 | ||
| Pneumonia | 159 (6.4%) | 19 (4.3%) | 0.12 | 57 (7.1%) | 17 (4.2%) | 0.065 |
| Missing data | 4 | 0 | 4 | 0 | ||
| DSWI | 30 (1.2%) | 4 (0.9%) | 0.8 | 9 (1.1%) | 4 (1.0%) | > 0.9 |
| Missing data | 1 | 0 | 1 | 0 | ||
| Delirium | 255 (10%) | 42 (9.6%) | 0.7 | 95 (12%) | 37 (9.2%) | 0.2 |
| Missing data | 5 | 0 | 4 | 0 | ||
| Atrial fibrillation | 987 (41%) | 162 (39%) | 0.4 | 335 (43%) | 153 (40%) | 0.3 |
| Missing data | 90 | 21 | 27 | 19 | ||
| Dialysis | 26 (1.0%) | 3 (0.7%) | 0.6 | 13 (1.6%) | 2 (0.5%) | 0.2 |
| Missing data | 4 | 2 | 4 | 2 | ||
| IABP | 27 (1.1%) | 4 (0.9%) | > 0.9 | 8 (1.0%) | 3 (0.7%) | 0.8 |
| Missing data | 1 | 0 | 1 | 0 | ||
| TnT (ng/l) | 350 (240–530) | 100 (60–230) | < 0.001 | 350 (240–530) | 100 (60–224) | < 0.001 |
| Missing data | 871 | 83 | 226 | 67 | ||
| Creatinine (μmol/l) | 89 (76–107) | 87 (73–108) | 0.081 | 91 (77–111) | 86 (72–107) | 0.003 |
| Missing data | 241 | 11 | 7 | 0 | ||
| CRP (mg/l) | 200 (157–243) | 195 (146–255) | 0.6 | 202 (161–247) | 196 (148–255) | 0.4 |
| Missing data | 42 | 1 | 13 | 0 | ||
Data from the first 30-days after operation. Continuous variables are reported as median (Q1–Q3). Categorical variables are summarized as n (%). Between‑group comparisons were conducted using the Wilcoxon rank‑sum test for continuous variables and either Pearson’s chi‑square test or Fisher’s exact test for categorical variables, where appropriate
ONCAB On-pump coronary artery bypass grafting, OPCAB Off-pump coronary artery bypass grafting, LOS length of hospital stay, ICU intensive care unit, TIA Transient ischaemic attack, AMI Acute myocardial infarction, DSWI Deep sternal wound infection, IABP Intra-aortic balloon pump, TnT Troponin-T level, CRP C-reactive protein
The median follow-up time was 11.5 years in the ONCAB group (IQR 7.7–16.3, maximum 23.9 years) and 10.9 years in the OPCAB group (IQR 7.3–14.0 years, maximum 23.8 years). The median survival time was 12.4 years (95% CI 11.9–13.2) in the ONCAB group and 12.6 years (95% CI 11.5–13.8) in the OPCAB group. Overall survival did not differ significantly between the groups (log-rank test, p = 0.51) (Fig. 2). In the Cox proportional hazards model, the HR for all-cause mortality was 0.95 (95% CI 0.82–1.11), with ONCAB as the reference group (Table 4). The Schoenfeld residual test indicated evidence of non-proportionality for operative technique, driven mainly by the end of follow-up when few patients remained at risk (p = 0.0225, supplementary figure). Cardiovascular death occurred in 271 ONCAB patients (33.7%) and in 106 OPCAB patients (26.4%) during follow-up. The cumulative incidence of cardiovascular death at 10 years was 18% (95% CI 16%–21%) in the ONCAB group and 19% (95% CI 15%–23%) in the OPCAB group. No significant difference in the cumulative incidence of cardiovascular death was observed between the groups (Gray’s test, p-value = 0.3). The sub-distribution HR for cardiovascular death was 0.85 (95% CI 0.68–1.07) with ONCAB as the reference (Fig. 3).
Fig. 2.
Survival
Table 4.
Univariate cox proportional hazards models for death as the outcome
| Characteristic | Unmatched | Matched | ||||||
|---|---|---|---|---|---|---|---|---|
| N | HR | 95% CI | p-value | N | HR | 95% CI | p-value | |
| 2,935 | 1,206 | |||||||
| ONCAB | — | — | — | — | ||||
| OPCAB | 1.05 | 0.92–1.20 | 0.5 | 0.95 | 0.82–1.11 | 0.5 | ||
| Sex | 2,935 | 1,206 | ||||||
| Male | — | — | — | — | ||||
| Female | 1.21 | 1.09–1.34 | < 0.001 | 1.24 | 1.06–1.45 | 0.007 | ||
| CCS | 2,791 | 1,157 | ||||||
| 1 | — | — | — | — | ||||
| 2 | 0.99 | 0.75–1.30 | > 0.9 | 0.94 | 0.65–1.37 | 0.8 | ||
| 3 | 1.31 | 1.00–1.71 | 0.047 | 1.34 | 0.93–1.92 | 0.12 | ||
| 4 | 1.56 | 1.19–2.05 | 0.001 | 1.65 | 1.14–2.39 | 0.008 | ||
| NYHA | 2,174 | 923 | ||||||
| 1 | — | — | — | — | ||||
| 2 | 0.92 | 0.69–1.24 | 0.6 | 0.93 | 0.63–1.38 | 0.7 | ||
| 3 | 1.36 | 1.02–1.80 | 0.033 | 1.45 | 0.99–2.13 | 0.057 | ||
| 4 | 1.52 | 1.14–2.04 | 0.004 | 1.76 | 1.18–2.62 | 0.005 | ||
| LV function | 2,799 | 1,206 | ||||||
| > 50% | — | — | — | — | ||||
| 31–50% | 1.41 | 1.28–1.56 | < 0.001 | 1.47 | 1.26–1.71 | < 0.001 | ||
| 21–30% | 2.85 | 2.34–3.47 | < 0.001 | 2.93 | 1.78–4.83 | < 0.001 | ||
| < 21% | 3.54 | 2.38–5.27 | < 0.001 | 2.78 | 1.32–5.87 | 0.007 | ||
| Diabetes | 2,935 | 1,206 | ||||||
| No | — | — | — | — | ||||
| Yes | 1.71 | 1.55–1.90 | < 0.001 | 1.56 | 1.32–1.84 | < 0.001 | ||
| COPD | 2,935 | 1,206 | ||||||
| No | — | — | — | — | ||||
| Yes | 1.83 | 1.60–2.09 | < 0.001 | 2.07 | 1.69–2.55 | < 0.001 | ||
| Poor mobility | 2,935 | 1,206 | ||||||
| No | — | — | — | — | ||||
| Yes | 1.97 | 1.43–2.70 | < 0.001 | 1.79 | 1.28–2.51 | < 0.001 | ||
| Extracardiac arteriopathy | 2,935 | 1,206 | ||||||
| No | — | — | — | — | ||||
| Yes | 2.24 | 1.96–2.57 | < 0.001 | 2.04 | 1.69–2.45 | < 0.001 | ||
| Renal impairment | 2,693 | 1,206 | ||||||
| Stage 1 | — | — | — | — | ||||
| Stage 2 | 1.58 | 1.41–1.77 | < 0.001 | 1.57 | 1.31–1.88 | < 0.001 | ||
| Stage 3 | 3.32 | 2.89–3.82 | < 0.001 | 3.08 | 2.49–3.81 | < 0.001 | ||
| Stage 4 | 11.4 | 7.34–17.8 | < 0.001 | 9.82 | 5.99–16.1 | < 0.001 | ||
| Stage 5 | 3.14 | 1.40–7.03 | 0.005 | 2.07 | 0.77–5.60 | 0.15 | ||
| Recent MI | 2,935 | 1,206 | ||||||
| No | — | — | — | — | ||||
| Yes | 1.30 | 1.18–1.43 | < 0.001 | 1.33 | 1.14–1.55 | < 0.001 | ||
Cox Proportional Hazards Models for Death as the Outcome
CI Confidence Interval, HR Hazard Ratio, ONCAB On-pump coronary artery bypass grafting, OPCAB Off-pump coronary artery bypass grafting, NYHA The New York Heart association classification, CCS The Canadien Cardiovascular Society classification, COPD Chronic obstructive pulmonary disease, Recent MI Myocardial infarction during the previous 90 days
Fig. 3.
Cardiovascular mortality
Discussion
In this single-centre propensity score-matched analysis with up to 20 years of follow-up, no significant difference was observed in either all-cause mortality or early post-operative complications between patients undergoing on-pump and off-pump CABG. Propensity score matching was used to mitigate confounding factors by indication and aimed to isolate the independent effect of surgical technique on survival. Our study adds to the existing literature by providing data from a real-world cohort, extending beyond the follow-up of most prior randomised trials [4, 9, 15, 16]. Furthermore, our findings suggest that equal long-term outcomes for both ONCAB and OPCAB can be achieved in a low-volume centre.
Our findings are consistent with the longest randomised trials reporting up to 20 years of follow-up [24], as well as a meta-analysis of 10-year follow-up data [20]. Large, registry-based and single-centre series have associated OPCAB with both similar and worse long-term survival [7, 8, 12, 21]. Both observational studies and randomised trials have suggested inferior outcomes in OPCAB technique, particularly in lower-volume settings [9, 15, 16]. Incomplete revascularisation, which is more frequent with OPCAB, has been identified as a key determinant of inferior long-term survival [9]. High-volume OPCAB centres have demonstrated equal long-term outcomes between the techniques when anastomotic quality and graft number are optimised [7, 21]. Our study shows that similar principles may be applicable in a smaller unit as well.
The theoretical advantages of OPCAB relate primarily to the avoidance of CPB-related systemic inflammation, embolic burden and end-organ dysfunction [3, 4]. These benefits may be particularly relevant in high-risk patients, such as the elderly or those with major aortic calcification or severe comorbidities. However, off-pump surgery is not suitable for all patients, as it presents technical challenges with potential trade-offs in graft number, target selection, or anastomotic quality [5, 6, 8]. This was observed in our study as the lower number of anastomoses conducted in the OPCAB patients. Also, differences were not observed in post-operative complications between the techniques. This finding may contradict the theoretical harm of CPB but may have resulted from optimal and individualised selection of patients to different procedures by the Heart Team.
Taken together with prior randomised and observational evidence, our findings support OPCAB as a viable alternative long-term revascularisation strategy when performed by experienced surgeons for appropriately selected patients. Attention should be paid to graft selection, distal anastomotic quality and completeness of revascularisation. The choice between ONCAB and OPCAB is then guided by a Heart Team-based approach considering individual patient risk profiles as well as coronary and aortic anatomy. Future research should focus on identifying patient subgroups most likely to benefit from OPCAB and reassessing the relative impact of surgical technique in the context of modern CPB and surgical techniques. Further studies could focus on how new surgical findings could be paired with other modern medical therapies addressing coronary artery disease.
Several limitations should be acknowledged. Firstly, this was a single-centre, observational study. Despite the use of propensity score matching, residual confounding cannot be excluded. Unmeasured factors such as coronary target quality, frailty, surgeon-specific decision-making, surgeon experience and patient selection may have influenced both treatment allocation and outcomes. These factors may have led to treating easier anatomies with OPCAB. Alternatively, aortic calcification may have been overrepresented in the OPCAB group. SYNTAX score or other standardised methods for describing the number of lesions were not available for our analyses. This makes it more challenging to address whether complete revascularisation was achieved. Secondly, surgical practise evolved over the prolonged inclusion period and included advances in medical therapy, graft selection and perioperative management. Although matching reduced baseline imbalances, secular trends may still have affected outcomes. Exclusion of patients with missing mortality or covariate data may have introduced selection bias, and residual bias related to missing data cannot be excluded. Thirdly, our findings may not be generalisable to centres with different case-mix profiles or operative expertise. The principal strengths of our study include the comprehensive long-term follow-up obtained from the national registry and the extensive, prospectively maintained institutional cardiac surgery database.
In conclusion, survival up to 20 years was comparable between ONCAB and OPCAB in this single-centre, low-volume setting.
Supplementary Information
Acknowledgements
We thank all our colleagues from the Vaasa Heart Team for their work throughout the study years and for making this study possible.
AI disclosure
ChatGPT 5.2 (OpenAI) was used for language editing to improve grammar and clarity.
Authors’ contributions
AP: Conceptualization, methodology, writing the original draft, data validation. PR: Writing, review and editing, administrative support. JOW: Writing, review and editing, data acquisition, conceptualization and administration of the institutional cardiac surgery registry. MH: Formal analysis, data validation, methodological oversight, writing the original draft. All authors contributed to the interpretation of the data, critically revised the manuscript, approved the final version for publication, and agree to be accountable for all aspects of the work, ensuring the accuracy and integrity of the published material.
Funding
Open Access funding provided by University of Oulu (including Oulu University Hospital). We would like to thank the Finnish State Research Funding and Kirsti & Tor Johansson’s Foundation for financial support towards this project.
Data availability
The data supporting the findings of this study are held at Vaasa Central Hospital. These restricted data are not publicly available and were used under license for the current study. These data are, however, available from the authors upon reasonable request and with permission of the Finnish Social and Health Data Permit Authority Findata.
Declarations
Ethics approval and consent to participate
The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The reported research has been conducted in an ethical and responsible manner and is in full compliance with all relevant codes of experimentation and legislation. The patient right to a separate written permission was waived by the ethical board’s decision due to the retrospective nature of the study. Additional approval from a separate ethics board was not required for this retrospective study. Study permission was received from the local ethics board of the Vaasa Central Hospital: ÖVPH/1192/13.01/2025. Clinical trial number: not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.GBD 2021 Causes of Death Collaborators. Global burden of 288 causes of death and life expectancy decomposition in 204 countries and territories and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2021;403:2100–32. 10.1016/S0140-6736(24)00367-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Weintraub WS, Clements SD, Crisco LV-T, Guyton RA, Craver JM, Jones EL, et al. Twenty-year survival after coronary artery surgery: an institutional perspective from Emory University. Circulation. 2003;107:1271–7. 10.1161/01.cir.0000053642.34528.d9. [DOI] [PubMed] [Google Scholar]
- 3.Raja SG. Two decades of off-pump coronary artery bypass surgery: Harefield experience. J Thorac Dis. 2016;8 Suppl 10:S824–8. 10.21037/jtd.2016.10.103. [DOI] [PMC free article] [PubMed]
- 4.Møller CH, Steinbrüchel DA. Off-pump versus on-pump coronary artery bypass grafting. Curr Cardiol Rep. 2014;16:455. 10.1007/s11886-013-0455-2. [DOI] [PubMed] [Google Scholar]
- 5.Filardo G, Hamman BL, da Graca B, Sass DM, Machala NJ, Ismail S, et al. Efficacy and effectiveness of on- versus off-pump coronary artery bypass grafting: A meta-analysis of mortality and survival. J Thorac Cardiovasc Surg. 2018;155:172-179.e5. 10.1016/j.jtcvs.2017.08.026. [DOI] [PubMed] [Google Scholar]
- 6.Zhou Z, Fu G, Feng K, Huang S, Chen G, Liang M, et al. Randomized evidence on graft patency after off-pump versus on-pump coronary artery bypass grafting: an updated meta-analysis. Int J Surg. 2022;98:106212. 10.1016/j.ijsu.2021.106212. [DOI] [PubMed] [Google Scholar]
- 7.Matkovic M, Tutus V, Bilbija I, Milin Lazovic J, Savic M, Cubrilo M, et al. Long term outcomes of the off-pump and on-pump coronary artery bypass grafting in a high-volume center. Sci Rep. 2019;9:8567. 10.1038/s41598-019-45093-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Benedetto U, Altman DG, Gerry S, Gray A, Lees B, Flather M, et al. Off-pump versus on-pump coronary artery bypass grafting: insights from the arterial revascularization trial. J Thorac Cardiovasc Surg. 2018;155:1545-1553.e7. 10.1016/j.jtcvs.2017.10.135. [DOI] [PubMed] [Google Scholar]
- 9.Benedetto U, Caputo M, Patel NN, Fiorentino F, Bryan A, Angelini GD. Long-term survival after off-pump versus on-pump coronary artery bypass graft surgery. Does completeness of revascularization play a role? Int J Cardiol. 2017;246:32–6. 10.1016/j.ijcard.2017.04.087. [DOI] [PubMed] [Google Scholar]
- 10.Neumann F-J, Sousa-Uva M, Ahlsson A, Alfonso F, Banning AP, Benedetto U, et al. 2018 ESC/EACTS Guidelines on myocardial revascularization. Eur Heart J. 2019;40:87–165. 10.1093/eurheartj/ehy394. [DOI] [PubMed] [Google Scholar]
- 11.Lawton JS, Tamis-Holland JE, Bangalore S, Bates ER, Beckie TM, Bischoff JM, et al. 2021 ACC/AHA/SCAI guideline for coronary artery revascularization: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2021;145:e18-114. 10.1161/CIR.0000000000001038. [DOI] [PubMed] [Google Scholar]
- 12.Squiers JJ, Schaffer JM, Banwait JK, Ryan WH, Mack MJ, DiMaio JM. Long-term survival after on-pump and off-pump coronary artery bypass grafting. Ann Thorac Surg. 2022;113:1943–52. 10.1016/j.athoracsur.2021.07.037. [DOI] [PubMed] [Google Scholar]
- 13.Dalén M, Ivert T, Holzmann MJ, Sartipy U. Long-term survival after off-pump coronary artery bypass surgery: a Swedish nationwide cohort study. Ann Thorac Surg. 2013;96:2054–60. 10.1016/j.athoracsur.2013.07.014. [DOI] [PubMed] [Google Scholar]
- 14.Shaefi S, Mittel A, Loberman D, Ramakrishna H. Off-Pump Versus On-Pump Coronary Artery Bypass Grafting-A Systematic Review and Analysis of Clinical Outcomes. J Cardiothorac Vasc Anesth. 2019;33:232–44. 10.1053/j.jvca.2018.04.012. [DOI] [PubMed] [Google Scholar]
- 15.Shroyer AL, Grover FL, Hattler B, Collins JF, McDonald GO, Kozora E, et al. On-pump versus off-pump coronary-artery bypass surgery. N Engl J Med. 2009;361:1827–37. 10.1056/NEJMoa0902905. [DOI] [PubMed] [Google Scholar]
- 16.Lamy A, Devereaux PJ, Prabhakaran D, Taggart DP, Hu S, Straka Z, et al. Five-Year Outcomes after Off-Pump or On-Pump Coronary-Artery Bypass Grafting. N Engl J Med. 2016;375:2359–68. 10.1056/NEJMoa1601564. [DOI] [PubMed] [Google Scholar]
- 17.Benedetto U, Lau C, Caputo M, Kim L, Feldman DN, Ohmes LB, et al. Comparison of outcomes for off-pump versus on-pump coronary artery bypass grafting in low-volume and high-volume centers and by low-volume and high-volume surgeons. Am J Cardiol. 2018;121:552–7. 10.1016/j.amjcard.2017.11.035. [DOI] [PubMed] [Google Scholar]
- 18.Diegeler A, Börgermann J, Kappert U, Breuer M, Böning A, Ursulescu A, et al. Off-pump versus on-pump coronary-artery bypass grafting in elderly patients. N Engl J Med. 2013;368:1189–98. 10.1056/NEJMoa1211666. [DOI] [PubMed] [Google Scholar]
- 19.Cappellaro AP, de Almeida LFC, Pinto ML, Martins MAB, Sousa AGE, Gadelha JG, et al. Off-pump versus on-pump coronary artery bypass grafting in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis. Gen Thorac Cardiovasc Surg. 2025;73:201–8. 10.1007/s11748-025-02116-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Comanici M, Bulut HI, Raja SG. 10-Year Mortality of Off-Pump Versus On-Pump Coronary Artery Bypass Grafting: An Updated Systematic Review, Meta-Analysis, and Meta-Regression. Am J Cardiol. 2024;219:77–84. 10.1016/j.amjcard.2024.03.019. [DOI] [PubMed] [Google Scholar]
- 21.Kim K-B, Choi JW, Oh SJ, Hwang HY, Kim JS, Choi J-S, et al. Twenty-Year Experience With Off-Pump Coronary Artery Bypass Grafting and Early Postoperative Angiography. Ann Thorac Surg. 2020;109:1112–9. 10.1016/j.athoracsur.2019.07.053. [DOI] [PubMed] [Google Scholar]
- 22.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;147:573–7. 10.7326/0003-4819-147-8-200710160-00010. [DOI] [PubMed] [Google Scholar]
- 23.Nashef SAM, Roques F, Sharples LD, Nilsson J, Smith C, Goldstone AR, et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41:734–44. 10.1093/ejcts/ezs043. [DOI] [PubMed] [Google Scholar]
- 24.Koop Y, Nathoe H, Bots M, Grobbee DE, Timmermans M, Wimmers RH, et al. Octopus follow-up: 20 year prognosis in patients randomized to on-pump CABG, off-pump CABG or PCI. Int J Cardiol. 2024;414:132426. 10.1016/j.ijcard.2024.132426. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting the findings of this study are held at Vaasa Central Hospital. These restricted data are not publicly available and were used under license for the current study. These data are, however, available from the authors upon reasonable request and with permission of the Finnish Social and Health Data Permit Authority Findata.