From 1990 to 2000, the number of individuals in the United States of age 80 years or older (octogenarians) increased from 7.0 to 9.2 million. It has been projected that the prevalence of this age group in the United States will number 15 million by the year 2025.1
Octogenarians undergoing coronary surgery typically present with an increased incidence of comorbid conditions and generally exhibit higher rates of major adverse events such as stroke, reoperation for bleeding, sepsis, and renal and respiratory failure in comparison with younger patients.2 Data have suggested that an octogenarian has more than three times the risk of dying after bypass surgery than a similar 50-year-old patient.3
There is lingering uncertainty over whether off-pump coronary artery bypass grafting (OPCAB) should be considered a superior alternative to on-pump CAB (ONCAB).4 Although some authors have been hesitant to advocate for OPCAB in this high-risk patient cohort,5,6 others have demonstrated improvements in mortality and morbidity, which has begun to facilitate the utilization of the OPCAB technique for octogenarians.7–10
The purpose of this study was to examine the difference in short- and long-term outcomes of death, stroke, myocardial infarction (MI), and major adverse cardiac events (MACE, the composite of death, stroke, and MI) in octogenarians undergoing OPCAB versus ONCAB.
MATERIALS AND METHODS
This study was implemented in compliance with the Health Insurance Portability and Accountability Act and the Declaration of Helsinki and was approved by the institutional review board of our institution. The Society of Thoracic Surgery (STS) adult cardiac database was searched for all patients who underwent isolated OPCAB or ONCAB at our hospitals between January 1996 and September 2008. Medical records of the study's cohort of 937 consecutive octogenarian patients included demographics, preexisting diseases, surgeon identity, operative strategy, and clinical outcomes.
Demographic and Preoperative Data
Standard STS definitions for preoperative risk factors were used when gathering and classifying data from the STS database (Table 1). Statistics were recorded into two groups representing OPCAB and ONCAB cohorts. Preoperative demographics accounted for on all patients included age, gender, congestive heart failure, previous MI, angina, previous cerebrovascular accident, cerebrovascular disease, peripheral vascular disease, chronic lung disease, diabetes mellitus, hypertension, infectious endocarditis, renal failure, dialysis, and the STS predicted risk of mortality. Race was dichotomized into Caucasian or non-Caucasian.
Categories missing data, assumed to be missing at random, included ejection fraction (n = 110, 11.3%), race (n = 71, 0.07%), New York Heart Association classification (n = 425, 43.7%), last creatinine level (n = 194, 19.9%), and number of diseased vessels (n = 69, 7.1%). All data on critical outcomes of interest in this study (surgery type and age) were 100% complete. A multiple imputation algorithm was used to impute missing values so that the whole sample could be analyzed. This was not done in an effort to recreate or emulate the true values; rather, the goal of the multiple imputation was to avoid selection bias that can occur by deleting cases with missing variables of interest. Five data sets were imputed, and estimates from the five data sets were combined using statistical methods.
Each patient considered in this study underwent either an ONCAB or OPCAB procedure conducted by 1 of 17 faculty surgeons who varied in their utilization of the off-pump technique. All patients were approached via a standard median sternotomy for cardiac access, and incision and closure techniques were similar in all cases. After pericardiotomy, patients prepped for ONCAB who were found to have a severely atheromatous ascending aorta (grade ≥3) by epiaortic ultrasound scanning were converted to OPCAB to avoid embolic complications resulting from aortic manipulation. Conventional ONCAB was performed with standard techniques, utilizing roller head pumps, membrane oxygenator, cardiotomy suction, arterial filter, cold antegrade and retrograde blood cardioplegia, and moderate systemic hypothermia (32–34°C). Patients who were converted intraoperatively from OPCAB to ONCAB or from ONCAB to OPCAB were entered into the database and analyzed according to the operation they ultimately received. Conversion to ONCAB was not recorded in the STS database until 2002 in our institution and, thus, is not available for a large proportion of patients. The OPCAB procedure required the use of commercially available stabilizing and positioning systems for the construction of distal anastomoses. Further details regarding the OPCAB and ONCAB techniques are described elsewhere.11,12
Postdischarge mortality was assessed based on the Social Security Death Index, an online database service maintained by the United States Social Security Administration. Persons with a social security number who have died since 1963 and been reported to the Social Security Administration are listed on the Social Security Death Index. The specific variable of interest in this study was the number of days between the date of surgery and the date of mortality. The cause of death was not available and, therefore, was not taken into account for the purpose of this study. Patients still alive on June 30, 2008, were considered censored in survival analyses.
Data Management and Statistical Analysis
Patients were classified based primarily on surgery type (ONCAB or OPCAB). Propensity scores, described in detail by Blackstone13 and D'Agostino14 were used to correct for potential selection bias based on 29 preoperative risk factors. An individual patient's propensity score is his or her probability of receiving OPCAB. Risk factors included surgeon identity, the year of surgery, and various clinical abnormalities (Table 1). Age, the most reliable risk factor for long-term survival, was not included in the propensity analysis and was instead accounted for directly in the final survival regression modeling.
To statistically evaluate the effects of surgery type on in-hospital mortality, a multivariable logistic regression model was constructed. The primary variable under consideration was surgery type (OPCAB or ONCAB). The model effects were adjusted using the propensity score. Adjusted odds ratios (AORs) associated with surgery type, along with 95% confidence intervals (CIs), were also computed.
Kaplan–Meier curves were generated that provide survival estimates at postoperative points 1-, 3-, 5-, and 10-year intervals. Differences between the surgery types were determined by log-rank tests. These estimates included operative deaths.
Adjusted long-term survival comparisons were made using Cox proportional hazards regression to model the instantaneous hazard of death as a function of surgery type, age, and the propensity score. The proportional hazards assumption was verified via a correlation analysis of the Schoenfeld residuals and ranked follow-up time. Adjusted hazard ratios (HRs) were generated for surgery type, along with 95% CI.
The data were managed and analyzed using SAS version 9.2 (Cary, NC). Unadjusted Table 1 comparisons were performed with χ2 tests and two-sample t tests for categorical and numerical predictors, respectively. All statistical tests were two sided using P = 0.05 level of significance.
The preoperative demographics, clinical characteristics, and risk factors for the 937 patients comprising the study population are represented in Table 1. Patients in the OPCAB cohort tended to be slightly older than ONCAB patients (82.9 vs 82.3; P = 0.003) and had a higher incidence of congestive heart failure (26.1% vs 20.4%; P = 0.04). The two groups differed significantly in the status of their operation with a higher percentage of OPCAB cases done urgently (23.9% vs 9.1%; P < 0.001) whereas more emergent cases were performed using ONCAB (8.8% vs 1.9%; P < 0.001).
When compared with the ONCAB cohort, the OPCAB cohort had a higher incidence of the following comorbid diseases: congestive heart failure (26.1% vs 20.4%; P = 0.04), peripheral vascular disease (15.6% vs 6.1%; P < 0.001), chronic lung disease (11.5% vs 2.5%; P < 0.001), and hypertension (86.3% vs 77.3%; P < 0.001). STS predicted risk of mortality was statistically equivalent between the two groups (5.3% vs 5.4%; P = 0.82).
The operative characteristics for all patients are represented in Table 2. The use of an intraoperative balloon pump was clinically higher in those undergoing ONCAB compared with the OPCAB cohort but was not statistically significant (P = 0.09). Body mass index was similar between groups.
Short-term postoperative outcomes (Table 3) revealed a longer mean intensive care unit stay in the ONCAB group (121.9 hours vs 75.6 hours; P = 0.016) compared with the OPCAB group. The ONCAB group also had a substantially higher rate of in-hospital mortality (9.3% vs 2.8%; P < 0.001). Incidence of the combined endpoint of MACE was twice as high in the ONCAB group (12.6% vs 5.2%; P < 0.001). The postoperative survival at 1, 3, 5, and 10-year time points approached each other over time (Fig. 1) and were not statistically significant in their differences (P = 0.22, Table 4). Adjusted survival was not different between the groups (OPCAB vs ONCAB HR = 0.77, 95% CI: 0.54–1.09, P = 0.14).
Propensity adjusted in-hospital outcomes by surgery type are listed in Table 5. The statistically significantly higher in-hospital mortality suffered by ONCAB patients persisted (9.32% vs 2.78%; P = 0.003) when compared with OPCAB patients. In addition, the incidence of ONCAB postoperative cerebrovascular accident was significantly higher when compared with the OPCAB cohort (3.78% vs 2.22%; P = 0.018). Finally, the overall incidence of MACE in the ONCAB group was significantly higher than in the OPCAB group (12.59% vs 5.19%; P < 0.001). OPCAB reduces the odds of in-hospital mortality by 75%, stroke by 74%, and MACE by 73%.
In this study, as in all previous studies from our institution, patients were noted ONCAB or OPCAB by the ultimate procedure performed. A substudy for the patients from 2002 forward (during which conversions are recorded) using the intention-to-treat approach revealed an AOR estimated at 0.33, only slightly larger than the 0.25 estimate that we have reported herein for the entire cohort. The conversion rate from OPCAB to ONCAB since 2002 was 12 of 416 patients (2.9%).
The results of this retrospective, propensity-adjusted study suggest that OPCAB, when applied to octogenarians, may significantly improve postoperative outcomes when compared with ONCAB. Our patients in the OPCAB cohort demonstrated statistically significant decreases of in-hospital mortality, stroke, and MACE compared with patients who underwent ONCAB. Findings from previous, smaller studies of octogenarians and OPCAB mirror these results, both for decreased in-hospital stay15 and stroke.7,15 These 2 studies were among 14 articles included in a 2006 meta-analysis of ONCAB versus OPCAB in the elderly.16 The authors specifically evaluated results for the two different strategies among patients older than 70 years. They found that OPCAB patients had an overall decreased incidence of death (OR: 0.48, 95% CI: 0.28–0.84). This effect was even more pronounced when the subset of octogenarians was examined (OR: 0.26, 95% CI: 0.12–0.57).
Despite multiple studies documenting the improvement of post-CAB outcomes in the absence of cardiopulmonary bypass (CPB),2,9,17–21 widespread adoption of the procedure has been gradual with OPCABs representing 20.4% of all CABs performed in the United States in 2007.22 The recent publication of results from the Randomized On/Off Bypass trial has added to the ongoing debate between proponents for ONCAB versus OPCAB.4
A large, prospective, randomized study involving 18 centers, the Randomized On/Off Bypass trial, examined results of over 2200 patients undergoing primary CABG using either ONCAB or OPCAB. Although the authors found no significant differences in short-term outcomes between the two groups, there were differences at 1 year. Specifically, they demonstrated that OPCAB patients were at a statistically significant increased risk for graft failure at 1 year as well as at an increased risk of death, nonfatal MI, or a need for repeat revascularization (9.9% vs 7.4%). Furthermore, they demonstrated a higher proportion of incomplete revascularization (fewer grafts completed than planned) among OPCAB patients (17.8% vs 11.1%).4 Although the results of this large trial are an important contribution to the existing literature, one must be careful with their interpretation. The study population was almost entirely male (>99%), relatively young (mean age, 63 years for OPCAB and 62.5 years or ONCAB), with a low predicted risk of mortality (1.9% OPCAB, 1.8% ONCAB), a group that tends to tolerate ONCAB very well.
A recent study underscores the fact that patients at higher risk are the ones most likely to benefit the most from an OPCAB approach.23 Using the predicted risk of mortality formula (PROM) used by the Society of Thoracic Surgeons, patients undergoing CABG were stratified according to their PROM and results compared for ONCAB versus OPCAB groups. Although there was no significant mortality difference for the lower-risk patients, those in the OPCAB group demonstrated a significant mortality benefit as the PROM increased above 2.5% to 3%.
A common criticism of the OPCAB technique is that patients with increasing numbers of diseased vessels are usually subject to CPB and these cases are inherently more complex and subject to a higher risk of adverse outcomes. A study conducted in 2008 explored this contention retrospectively on 12,812 consecutive adult patients undergoing primary, isolated CAB.12 The study used the ratio of the number of grafts constructed to the number of diseased vessels as an index of the completeness of revascularization. HR analysis concluded that higher age and lower index of the completeness of revascularization significantly worsened postoperative mortality whereas the method or extent of revascularization did not. In other words, the number of diseased vessels should not be considered a significant risk factor as long as there is completeness of revascularization. In 2006, 75% of patients receiving four to seven vessel grafts at our institution's hospitals did so without CPB.
For the most part, limitations of this study revolve around its retrospective, nonrandomized nature. Any retrospective study is subject to potentially confounding variables that are exceedingly difficult to incorporate even with advanced statistical models. Missing data, particularly for the number of diseased vessels and left ventricular ejection fraction, may also be viewed as shortcomings in risk factor analyses. Moreover, our institution's long-standing commitment to and experience with the OPCAB approach may make it difficult to apply these results to lower volume centers.
Another limitation was that patients were analyzed according to the procedure ultimately performed, not intention to treat. Although STS score and propensity matching accounts for much of the difference in expected mortality between groups, this may explain some of the increased mortality seen in ONCAB patients who were converted from OPCAB.
In summary, this represents a large cohort of patients comparing outcomes of OPCAB and ONCAB in the octogenarian population. The results suggest that potential benefits of OPCAB when applied to elderly patients include significant decreases of in-hospital mortality, stroke, and overall MACE. The clinical benefits derived from OPCAB when used by an experienced surgeon have been confirmed by many authors, and its use among octogenarians should be considered a vital component of any revascularization strategy involving this higher-risk and increasingly growing group of patients.
The authors thank staff members Kim Baio for project oversight, Jean Walker and Susan Joyce for data abstraction, and Deborah Canup for database management.
2. Boyd WD, Desai ND, Del Rizzo DF, et al. Off-pump surgery decreases postoperative complications and resource utilization in the elderly. Ann Thorac Surg
3. Alexander KP, Peterson ED. Coronary artery bypass grafting in the elderly. Am Heart J
. 1997;134(5 Pt 1):856–864.
4. Shroyer AL, Grover FL, Hattler B, et al. On-pump versus off-pump coronary-artery bypass surgery. N Engl J Med
5. Yacoub M. Off-pump coronary bypass surgery: in search of an identity. Circulation
6. Ricci M, Karamanoukian HL, Abraham R, et al. Stroke in octogenarians undergoing coronary artery surgery with and without cardiopulmonary bypass. Ann Thorac Surg
7. Hoff SJ, Ball SK, Coltharp WH, et al. Coronary artery bypass in patients 80 years and over: is off-pump the operation of choice? Ann Thorac Surg
8. Beauford RB, Goldstein DJ, Sardari FF, et al. Multivessel off-pump revascularization in octogenarians: early and midterm outcomes. Ann Thorac Surg
9. Al-Ruzzeh S, Ambler G, Asimakopoulos G, et al. Off-pump coronary artery bypass (OPCAB) surgery reduces risk-stratified morbidity and mortality: a United Kingdom multi-center comparative analysis of early clinical outcome. Circulation
. 2003;108(suppl 1):II1–II8.
10. Li Y, Zheng Z, Hu S. Early and long-term outcomes in the elderly: comparison between off-pump and on-pump techniques in 1191 patients undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg
11. Puskas JD, Williams WH, Duke PG, et al. Off-pump coronary artery bypass grafting provides complete revascularization with reduced myocardial injury, transfusion requirements, and length of stay: a prospective randomized comparison of two hundred unselected patients undergoing off-pump versus conventional coronary artery bypass grafting. J Thorac Cardiovasc Surg
12. Lattouf OM, Thourani VH, Kilgo PD, et al. Influence of on-pump versus off-pump techniques and completeness of revascularization on long-term survival after coronary artery bypass. Ann Thorac Surg
13. Blackstone EH. Comparing apples and oranges. J Thorac Cardiovasc Surg
14. D'Agostino RB Jr. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med
15. Demaria RG, Carrier M, Fortier S, et al. Reduced mortality and strokes with off-pump coronary artery bypass grafting surgery in octogenarians. Circulation
. 2002;106(12 suppl 1):I5–I10.
16. Panesar SS, Athanasiou T, Nair S, et al. Early outcomes in the elderly: a meta-analysis of 4921 patients undergoing coronary artery bypass grafting—comparison between off-pump and on-pump techniques. Heart
17. Cleveland JC Jr, Shroyer AL, Chen AY, et al. Off-pump coronary artery bypass grafting decreases risk-adjusted mortality and morbidity. Ann Thorac Surg
18. Athanasiou T, Aziz O, Mangoush O, et al. Do off-pump techniques reduce the incidence of postoperative atrial fibrillation in elderly patients undergoing coronary artery bypass grafting? Ann Thorac Surg
19. D'Alfonso A, Mariani MA, Amerini A, et al. Off-pump coronary surgery improves in-hospital and early outcomes in octogenarians. Ital Heart J
20. Meco M, Biraghi T, Panisi P, et al. Aortocoronary bypass grafting in high-risk patients over 75 years. Propensity score analysis of on versus off-pump, early and midterm results. J Cardiovasc Surg (Torino)
21. Puskas JD, Kilgo PD, Lattouf OM, et al. Off-pump coronary bypass provides reduced mortality and morbidity and equivalent 10-year survival. Ann Thorac Surg
22. Society of Thoracic Surgeons National Adult Cardiac Database. Spring Report 2008
. Durham, NC: Duke Clinical Research Institute; June 2008.
23. Puskas JD, Thourani VH, Kilgo P, et al. Off-pump coronary artery bypass disproportionately benefits high-risk patients. Ann Thorac Surg
This interesting report from the pioneering group at Emory compares the short- and long-term outcomes in octogenarian patients undergoing off-pump (OPCAB) vs. on-pump (ONCAB) coronary bypass grafting. It represents one of the largest studies in the literature focused on OPCAB results in the elderly. They performed a propensity-adjusted retrospective review of 937 patients. OPCAB had a significant lower in-hospital mortality in this matched trial (2.8% vs. 9.3%, P = .0007. There was no difference in long-term mortality. In this analysis, patients were grouped according to the ultimate procedure performed and not by intention to treat. However, their conversion rate was only 2.9%. Even when grouped by intention to treat, there still was a benefit in terms of mortality to OPCAB in this group of elderly patients. The incidence of major adverse cardiac events was twice as high in the ONCAB group (12.6% vs 5.2%, p<0.001).
The results of this study suggest the potential benefits of OPCAB in elderly patients. The main limitation of this study is its retrospective and non-randomized nature. This institution also has a long-standing commitment to OPCAB and their results may not be reproducible in lower-volume and less dedicated centers. However, this is another outstanding contribution from the Emory group and suggests that in the hands of experienced surgeons, OPCAB is an important component of the revascularization strategy in higher risk elderly patients.