ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Ann Thorac Surg 2012;93:1167-1172. doi:10.1016/j.athoracsur.2011.12.033
© 2012 The Society of Thoracic Surgeons
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Li, Z.
Right arrow Articles by Young, J. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Li, Z.
Right arrow Articles by Young, J. N.
Related Collections
Right arrow Coronary disease
Right arrowRelated Article

Original Articles: Adult Cardiac

Coronary Artery Bypass Operations for Elderly Patients in California, 2003 to 2008

Zhongmin Li, PhDa,*, Ezra A. Amsterdam, MDa, Khung Keong Yeo, MDa, Holly Hoegh, PhDb, Geeta Mahendra, MSa, W. Douglas Boyd, MDa, J. Nilas Young, MDa

a University of California, Davis Medical Center, Sacramento, California
b California Office of Statewide Health Planning and Development, Sacramento, California

Accepted for publication December 12, 2011.

Abbreviations and Acronyms CABG = coronary artery bypass grafting; CCB = conventional on-pump coronary bypass; CCORP = California Coronary Artery Bypass Graft Outcomes Reporting Program; CI = confidence interval; CVA = cerebrovascular accident; MI = myocardial infarction; O/E = observed-to-expected ratio; OPCAB = off-pump coronary artery bypass grafting; OR = odds ratio; OSHPD = California Office of Statewide Health Planning and Development; PCI = percutaneous coronary intervention; VF = ventricular fibrillation; Vfib = ventricular fibrillation; VT = ventricular tachycardia; Vtach = ventricular tachycardia

* Address correspondence to Dr Li, Department of Internal Medicine, University of California, Davis Medical Center, 4150 V St, Ste 2400, Sacramento, CA 95817 (Email: zhongmin.li{at}ucdmc.ucdavis.edu).


   Abstract
Top
 Abstract
Introduction
 Material and Methods
 Results
 Comment
 References
 
Background: Coronary artery bypass grafting (CABG) is no longer rare for elderly patients. This study evaluates operative mortality and the effects of off-pump CABG (OPCAB) on mortality for elderly Californians between 2003 and 2008.

Methods: All isolated CABGs in California for 2003 to 2008 were classified into cohorts by age: (1) younger than 75, (2) 75 to 84, and (3) 85 or older. Multivariable logistic regression models were developed for operative mortality. Trend analyses for observed and predicted mortality, and observed-to-expected mortality ratios were performed. The "recycled predictions" method was used to assess the effect of OPCAB on operative mortality.

Results: Among 101,710 isolated CABGs between 2003 and 2008, 22.0% were in cohort 2 and 2.3% were in cohort 3. Predicted mortality was unchanged for cohorts 2 and 3 (all p > 0.05), but observed-to-expected mortality ratios declined from 0.958 to 0.633 for cohort 2 (p = 0.021) and from 1.027 to 0.965 for cohort 3 (p = 0.168). The proportion of OPCAB for patients aged 75 years or older increased from 25.0% to 29.1% between 2003 and 2008. The adjusted odds ratio for operative mortality for OPCAB in patients aged 75 years or older was 0.752 (95% confidence interval, 0.650 to 0.871; p < 0.001) compared with on-pump CABG for the same age cohort.

Conclusions: In California, overall predicted mortality was unchanged for elderly patients between 2003 and 2008, but operative mortality significantly decreased for patients aged 75 to 84. Improvement for CABG patients aged 85 years or older was insignificant. The increase in the number of OPCAB patients was associated with decreased mortality for elderly patients.


   Introduction
Top
 Abstract
 Introduction
Material and Methods
 Results
 Comment
 References
 
Advanced age is an important risk factor for mortality and morbidity in patients undergoing coronary artery bypass grafting (CABG) operations. Concern regarding the increased operative risk of elderly patients is reflected in The Society of Thoracic Surgeons 2008 risk model. For example, the estimated odds ratio for operative mortality in patients undergoing isolated CABG is 4.7 in those aged 80 years compared with those aged 50 years [1]. However, with advances in medical care and increasing life expectancy, CABG is no longer uncommon in elderly patients with multivessel coronary artery disease. Thus, there have been a variety of reports of CABG outcomes in elderly patients, including octogenarians, nonagenarians, and centenarians [2–12]. The aim of this study was to investigate the trend in operative mortality of elderly patients who underwent isolated CABG between 2003 and 2008 in California and to explore the effect of off-pump bypass (OPCAB) operations on operative mortality for elderly patients.


   Material and Methods
Top
 Abstract
 Introduction
 Material and Methods
Results
 Comment
 References
 
This study was approved by the Committee for the Protection of Human Subjects of the California Health and Human Services Agency on June 3, 2011 (Project # 04–08–58).

Source
Data were obtained from the California Coronary Artery Bypass Graft Outcomes Reporting Program (CCORP) data registry on patients who underwent CABG operations between 2003 and 2008. Beginning in 2003, California state law has required that all state-licensed hospitals that perform isolated and nonisolated CABG report their cases to CCORP. Isolated CABG is defined as CABG performed without other major procedures during the same operation, including valve repair, carotid endarterectomy, or aortic, lung, or ventricular operations.

The CCORP clinical data registry uses a subset of data elements collected by The Society of Thoracic Surgeons for its Adult Cardiac Surgery Database. After data submission by hospitals, CCORP performs validation procedures, including verification of complete submissions using the statewide hospital patient discharge data, verification of deaths using the state's death registry, and an annual independent medical record audit of selected hospitals. It takes about 2 years from data being submitted by each hospital performing CABGs to the completion of data cleaning, on-site audit, and validation. The program produces hospital and surgeon-specific report cards based on risk-adjusted operative mortality rates. A detailed description of the CCORP data collection methods and analysis has been documented elsewhere [13].

All isolated CABG cases in the CCORP database between 2003 and 2008 were classified into three cohorts by age in years: (1) younger than 75, (2) 75 to 84, and (3) 85 or older. Operative mortality was defined as a patient who died at hospital discharge or after discharge but within 30 days of CABG. To investigate the effect of OPCAB on operative mortality in the elderly, we categorized those "combined" procedures—in which OPCAB procedures were converted to conventional coronary bypass (CCB) intraoperatively—as OPCAB based on the intention-to-treat principle.

Statistical Analyses
We performed the Cochran-Armitage trend test to compare historical change in case volume and observed operative mortality for patients by age cohorts between 2003 and 2008. To compare the patient demographic and clinical risk profile between 2003 and 2008, we used the {chi}2 test for categoric variables and Student t test for continuous variables. To avoid prediction bias, we developed a multivariable logistic regression model with 2-year combined data (2003 and 2004) and applied the model to each year between 2003 and 2008 for computation of predicted operative mortality for each patient. We computed mean expected mortality and the observed-to-expected (O/E) mortality ratios by age cohort for each year between 2003 and 2008 and used the general linear regression model to test the differences in observed mortality, expected mortality, and O/E mortality ratio for the elderly cohorts between 2003 and 2008.

We then added OPCAB to the multivariable risk model and used the "recycled predictions" method to compute and compare the mean risk-adjusted operative mortality rates between OPCAB and CCB for elderly patients. The method of recycled predictions produces the marginal effect that measures the expected instantaneous change in the dependent variable (ie, operative mortality) as a function of a change in a certain explanatory variable (ie, OPCAB vs CCB) while keeping all the other covariates constant [14, 15]. By using the recycled predictions method for the first run, we assumed CCB was performed in all elderly patients, and for the second run, we assumed OPCAB was performed. All differences in statistical analysis were considered significant if p was less than 0.05. All data analyses were conducted with SAS 9.2 software (SAS Institute, Cary, NC).


   Results
Top
 Abstract
 Introduction
 Material and Methods
 Results
Comment
 References
 
Among 101,710 patients who underwent isolated CABGs between 2003 and 2008 in California, 76,934 (75.6%) were in cohort 1 (age <75 years), 22,398 (22.0%) were in cohort 2 (age 75–84 years), and 2,378 (2.3%) were in cohort 3 (age ≥85 years). The annual volume of isolated CABG for all ages fell by 34.4% from 21,276 in 2003 to 13,957 in 2008, and the total number of elderly patients (aged ≥75 years) undergoing isolated CABG also decreased from 5,356 in 2003 to 3,264 in 2008, a 39% reduction over 6 years (Cochran-Armitage trend test p < 0.0001). However, the proportion of patients aged 85 or older among all isolated CABGs increased slightly, from 2.2% to 2.4% (p = 0.013), for the same period.

Cohorts 2 and 3 had a higher prevalence in 2008 than in 2003 for the following: age 85 or older, nonwhite race, hypertension, dialysis, cerebrovascular disease, diabetes, severe chronic lung disease, immunosuppressive treatment, atrial fibrillation, prior percutaneous coronary intervention, left main stenosis exceeding 75%, and moderate–severe mitral insufficiency (all p < 0.05). In 2008, however, elderly patients had a lower prevalence of emergency or salvage CABG, myocardial infarction (MI) within 24 hours of CABG, heart block, one or more prior cardiac operations, or cardiogenic shock (all p < 0.05; Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1 Patient Demographic and Clinical Profile for Patients Aged 75 Years and Older: 2003 vs 2008
 
A multivariable logistic regression model for predicting operative mortality was developed using 2003 and 2004 combined isolated CABGs (n = 40,377). The model included 24 demographic and clinical risk factors (Fig 1) and had a C statistic of 0.817, indicating good discrimination, and p = 0.145 for the Hosmer-Lemeshow test, indicating an acceptable data calibration. The risk model showed 19 factors were significant for operative mortality (all p < 0.05). Among these, body mass index of less than 18.5 kg/m2, emergency or salvage status, creatinine level, severe chronic lung disease, arrhythmia ventricular tachycardia/ventricular fibrillation, and cardiogenic shock were the most significant risk factors, with odds ratios exceeding 2.0. The risk model was validated with CCORP 2005 to 2008 data (n = 61,333) and had a C statistic of 0.810.


Figure 1
View larger version (5K):
[in this window]
[in a new window]

  		Fig 1. Multivariable logistic regression model for operative mortality. (Afib = atrial fibrillation; CVA = cerebrovascular accident; MI = myocardial infarction; PCI = percutaneous coronary intervention; preOp = preoperative; Sust = sustained; VF = ventricular fibrillation; VT = ventricular tachycardia.)

 
We also conducted data calibration tests with the Hosmer-Lemeshow partition method for "training" and "test" data sets. The results from two data sets showed none of the 10 risk groups (ie, deciles) had significantly fewer or more deaths than were predicted by the model. The multivariable logistic regression model showed the overall predicted/expected mortality for elderly patients was unchanged between 2003 and 2008, with cohort 2 from 5.40% to 5.45% (p = 0.996) and cohort 3 from 8.81% to 7.62% (p = 0.088). During this same interval, the observed operative mortality rate declined from 5.17% to 3.45% for cohort 2 (p = 0.001) and from 9.05% to 7.35% for cohort 3 (p = 0.993; Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2 Comparisons of Predicted and Observed Operative Mortality Between 2003 and 2008
 
Figure 2 presents the changes in O/E mortality ratios over time by age cohort. The O/E mortality ratio represents the patient risk-adjusted operative outcome: if the O/E ratio exceeds 1, the specified age cohort had a higher-than-expected mortality based on the patient case-mix; if the O/E ratio is less than 1, the age cohort had a lower than expected mortality. As shown in Figure 2, the O/E ratios declined between 2003 and 2008 for all age cohorts, which led to an overall significant decline in the O/E mortality ratio for the entire patient population (p = 0.014). However, among three age cohorts only the change over time for cohort 2 (age 75 to 84 years) was statistically significant (p = 0.021).


Figure 2
View larger version (4K):
[in this window]
[in a new window]

  		Fig 2. Trend of observed-to-expected (O/E) mortality ratios from 2003 to 2008 by age cohorts: < 75 years (diamonds), 75 to 84 (squares), ≥ 85 years (triangles), and all ages (circles).

 
Although the proportion of patients undergoing OPCAB among all isolated CABGs increased slightly from 23.1% in 2003 to 24.9% in 2008 (p = 0.212), this proportion increased from 25.0% to 29.1% for patients aged 75 years or older (p = 0.047). Compared with the patients aged 75 years or older who underwent CCB, the same age cohort who underwent OPCAB had a higher prevalence in older age, female sex, body mass index of less than 18.5 kg/m2, dialysis, peripheral vascular disease, cerebrovascular disease, chronic lung disease, prior percutaneous coronary intervention, and immunosuppressive treatment (all p < 0.05), but less prevalence in emergency or salvage procedures, recent myocardial infarction, cardiogenic shock, left ventricular ejection fraction of less than 0.40, and 3 or more diseased vessels (all p < 0.05). There were no differences in cerebrovascular accident, diabetes, heart failure, left main stenosis of 75% or more, prior CABG, and mitral insufficiency between OPCAB and CCB patients (all p > 0.05).

We then added OPCAB into the multivariable logistic regression model for operative mortality for different age cohorts. Table 3 presents comparisons of adjusted odds ratios of OPCAB vs CCB for operative mortality among different age cohorts. Although overall OPCAB was associated with a protective effect on operative mortality for the entire isolated CABG population in California from 2003 to 2008, the patients aged 75 or older benefitted the most from OPCAB on operative mortality (adjusted odds ratio, 0.752; 95% confidence interval, 0.650 to 0.871; p = 0.0001) vs patients younger than 75 (adjusted odds ratio. 0.840; 95% confidence interval, 0.734 to 0.964; p = 0.013). However, the protective effect on patients aged 85 years or older was inconclusive (p = 0.1615, Table 3).


View this table:
[in this window]
[in a new window]

  		Table 3 Adjusted Odds Ratios for Off-Pump Coronary Artery Bypass vs Conventional On-Pump Coronary Bypass in Multivariable Logistic Regression Models for Operative Mortality, a 2003 to 2008
 

   Comment
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
References
 
This study demonstrates a favorable trend between 2003 and 2008 in operative mortality for isolated CABG in elderly patients, especially in the cohort aged 75 to 84 years. Most prior studies compared elderly patients with younger cohorts in risk prevalence and in predicted and risk-adjusted mortality and morbidity and found that advanced age did not necessarily exclude elderly patients from cardiac operations if patients had relatively few risk factors [2, 3, 6].

However, there is a paucity of studies on the trend of CABG outcomes in elderly patients. We found that although the proportion of very elderly patients (>75 years) undergoing CABG in California decreased during the interval from 2003 to 2008, the overall predicted risk for mortality in this group was unchanged. The O/E mortality ratios for all age cohorts in California declined after 2003, but the decrease for the cohort aged 75 to 85 years was more pronounced. Compared with 2003, the percentage of elderly patients undergoing isolated CABG in 2008 was 1.2% lower in those with emergency status and 2.1% lower in those with myocardial infarction within 24 hours of operation. These findings are likely related to a variety of factors such as patient selection and preoperative, intraoperative, and postoperative management. More important, our data indicate that this beneficial outcome was not related to reduced operative risk because the predicted operative mortality in elderly patients undergoing isolated CABG was not lower in 2008 than in 2003 (all p > 0.05, Table 2).

A number of earlier studies found advanced age was a significant risk factor for operative mortality and complications after controlling for other risk factors [1–6]. Our study extended these prior findings by increasing the number of risk factors to 24 in the prediction model for CABG operative mortality. We found that controlling for all other risk factors, operative mortality increased by 5.7% in the multivariable logistic regression risk model for every additional year of age increase.

LaPar and colleagues [16] reported in 2011 that performance of CABG in octogenarian patients was safe and effective. They also noted that OPCAB conferred shorter postoperative ventilation time with equivalent mortality to conventional CABG. Our multivariable logistic regression risk model showed a 20% lower predicted risk for operative mortality for elderly patients who underwent OPCAB compared with those undergoing CCB. Our recycled predictions method and the comparison of observed mortality showed risk-adjusted mortality and observed mortality for OPCAB for the elderly were both less than the risk-adjusted mortality and observed mortality for CCB. Although we found that the protective effect of OPCAB on patients aged 85 years or older was inconclusive, our findings suggest that OPCAB was beneficial in elderly patients because it was associated with a significantly lower operative mortality than CCB.

In an analysis of approximately 10,000 CABG patients between 1999 and 2007 by Pierri and colleagues [17], the proportion of those aged 75 years or older increased from 17% in 1999 to 29% in 2005, while the percentage of patients aged 80 years or older increased threefold. As a consequence, there was an increased calculated surgical risk. Although our study found that the overall number of elderly patients undergoing CABG in California has decreased over time and the predicted mortality for elderly patients was unchanged between 2003 and 2008, the demographic aging process in the developed nations and advancing operative techniques may result in more elderly patients undergoing CABG in the future.

This report is a retrospective observational study with the inherent limitations of this method. Further, this study is limited to operative mortality as the only outcome measure and does not include postoperative complications or long-term clinical outcomes. However, CCORP started collection of data on postoperative complications in 2006, and additional years of data are needed to complete analysis for other outcome measures for elderly patients. We also acknowledge that data sets such as ours may not include all clinical elements affecting postoperative mortality. The study includes information on patients and surgeons from nonfederal California hospitals, thus limiting the generalizability of our results. Finally, the CCORP data did not specify the individual diseased coronary arteries and the site in the artery at which a graft was placed. The foregoing limitations notwithstanding, this study includes detailed data on important patient operative risk profiles that were used for an extensive analysis of our clinical findings.

In conclusion, the number of elderly patients undergoing CABG in California decreased from 2003 to 2008 but the overall predicted risk for mortality was unchanged. The observed mortality and O/E mortality ratio declined for the elderly, especially for patients aged 75 to 84 years during this interval. Finally, the proportion of elderly patients undergoing OPCAB significantly increased over time, and this method was associated with reduced odds of operative mortality.


   References
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
 References
 

  1. Shahian DM, O'Brien SM, Filardo G, et al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 1—coronary artery bypass grafting surgery Ann Thorac Surg 2009;88(1 Suppl):S2-S22.[Abstract/Free Full Text]
  2. Bridges CR, Edwards FH, Peterson ED, Coombs LP, Ferguson TB. Cardiac surgery in nonagenarians and centenarians J Am Coll Surg 2003;197:347-357.[Medline]
  3. Alexander KP, Anstrom KJ, Muhlbaier LH, et al. Outcomes of cardiac surgery in patients > or = 80 years: results from the National Cardiovascular Network J Am Coll Cardiol 2000;35:731-738.[Medline]
  4. Kolh P, Kerzmann A, Lahaye L, Gerard P, Limet R. Cardiac surgery in octogenarians; peri-operative outcome and long-term results Eur Heart J 2001;22:1235-1243.[Abstract/Free Full Text]
  5. Zingone B, Gatti G, Rauber E, et al. Early and late outcomes of cardiac surgery in octogenarians Ann Thorac Surg 2009;87:71-78.[Abstract/Free Full Text]
  6. Speziale G, Nasso G, Barattoni MC, et al. Operative and middle-term results of cardiac surgery in nonagenarians: a bridge toward routine practice Circulation 2010;121:208-213.[Abstract/Free Full Text]
  7. Guilfoyle MR, Drain AJ, Khan A, Ferguson J, Large SR, Nashef SA. Cardiac surgery in nonagenarians: single-centre series and review Gerontology 2010;56:378-384.[Medline]
  8. Peterson ED, Cowper PA, Jollis JG, et al. Outcomes of coronary artery bypass graft surgery in 24,461 patients aged 80 years or older Circulation 1995;92(9 Suppl):II85-II91.[Medline]
  9. Johnson WM, Smith JM, Woods SE, Hendy MP, Hiratzka LF. Cardiac surgery in octogenarians: does age alone influence outcomes? Arch Surg 2005;140:1089-1093.[Medline]
  10. Craver JM, Puskas JD, Weintraub WW, et al. 601 octogenarians undergoing cardiac surgery: outcome and comparison with younger age groups Ann Thorac Surg 1999;67:1104-1110.[Abstract/Free Full Text]
  11. Williams DB, Carrillo RG, Traad EA, et al. Determinants of operative mortality in octogenarians undergoing coronary bypass Ann Thorac Surg 1995;60:1038-1043.[Abstract/Free Full Text]
  12. Naughton C, Feneck RO, Roxburgh J. Early and late predictors of mortality following on-pump coronary artery bypass graft surgery in the elderly as compared to a younger population Eur J Cardiothorac Surg 2009;36:621-627.[Abstract/Free Full Text]
  13. California Office of Statewide Health Planning and Development The California report on coronary artery bypass graft surgery 2005–2006 hospital and surgeon data. Sacramento, CA: California Office of Statewide Health Planning and Development; March 2009.
  14. Greene WH. Econometric analysis. 3rd edition. Upper Saddle River, NJ: Prentice Hall; 1997. pp. 339-350.
  15. Graubard B, Edward L, Korn E. Predictive margins with survey data Biometrics 1999;55:652-659.[Medline]
  16. LaPar DJ, Bhamidipati CM, Reece TB, Cleveland JC, Kron IL, Ailawadi G. Is off-pump coronary artery bypass grafting superior to conventional bypass in octogenarians? J Thorac Cardiovasc Surg 2011;141:81-90.[Abstract/Free Full Text]
  17. Pierri MD, Capestro F, Zingaro C, Torracca L. The changing face of cardiac surgery patients: an insight into a Mediterranean region Eur J Cardiothorac Surg 2010;38:407-413.[Abstract/Free Full Text]

Related Article

Invited Commentary
Edward L. Hannan
Ann. Thorac. Surg. 2012 93: 1172-1173. [Extract] [Full Text] [PDF]



This article has been cited by other articles:


Home page
 Ann. Thorac. Surg.Home page
E. L. Hannan
Invited Commentary
Ann. Thorac. Surg., April 1, 2012; 93(4): 1172 - 1173.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Li, Z.
Right arrow Articles by Young, J. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Li, Z.
Right arrow Articles by Young, J. N.
Related Collections
Right arrow Coronary disease
Right arrowRelated Article

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
ANN THORAC SURG ASIAN CARDIOVASC THORAC ANN EUR J CARDIOTHORAC SURG
J THORAC CARDIOVASC SURG ICVTS ALL CTSNet JOURNALS