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


     

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 Hartz, A. J.
Right arrow Articles by Assa, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hartz, A. J.
Right arrow Articles by Assa, J.

Ann Thorac Surg 2000;69:829-833
© 2000 The Society of Thoracic Surgeons

Original Articles

Trends during 25 years of coronary artery bypass operation at St. Luke’s Medical Center in Milwaukee, Wisconsin

Arthur J. Hartz, MD, PhDa, Jack C. Manley, MDb, John A. Walker, MDb, Henry H. Gale, MDb, Zhixiong He, PhDa, Jacob Assa, MSa

a Department of Family Medicine, University of Iowa College of Medicine, Iowa City, Iowa, USA
b Department of Cardiology, St. Luke’s Medical Center, Milwaukee, Wisconsin, USA

Address reprint requests to Dr Hartz, Department of Family Medicine, University of Iowa College of Medicine, 01292-D PFP, Iowa City, IA 52242-1097
e-mail: arthur-hartz{at}uiowa.edu


   Abstract
Top
 Abstract
Introduction
 Material and methods
 Results
 Comment
 References
 
Background. Most studies of changes in coronary artery bypass graft (CABG) operations are from major academic institutions. The present study evaluated changes in CABG operations since 1968 in a community hospital.

Methods. The data were from the St. Luke’s Medical Center Cardiovascular Data Registry in Milwaukee, Wisconsin. Mortality rates, risk factors, overall patient risk, and surgical procedures were compared from 1968 to 1994.

Results. There was a dramatic decrease in 30-day mortality rates from 1968 to 1972. After 1976, mortality rates increased because of higher risk patients, but the mortality rate, adjusted for patient risk, continued to decline. Both internal mammary arteries and sequential grafts were widely used by 1972, followed by a decline in use until 1980, and then a steep increase in use from 1980 to the present.

Conclusions. This study provided evidence from a community hospital that the skills of the surgical teams improved first dramatically then gradually. The pattern of adapting new surgical techniques suggested that these techniques were critically evaluated for several years after they were introduced.


   Introduction
Top
 Abstract
 Introduction
Material and methods
 Results
 Comment
 References
 
Several studies have documented changes in mortality rates, patient characteristics, and surgical techniques in patients who had coronary artery disease bypass graft (CABG) operations [1–16]. Almost all of these studies were from major academic institutions. The studies differed with respect to the time period examined, the data elements compared, and the inclusion of patients with concomitant procedures or reoperations. The present study tracked changes in mortality rates, mortality risk, and surgical techniques at a single community hospital from 1968, a time almost at the inception of CABG operation, to 1994.


   Material and methods
Top
 Abstract
 Introduction
 Material and methods
Results
 Comment
 References
 
The data were obtained from the St. Luke’s Medical Center Cardiovascular Data Registry in Milwaukee, Wisconsin. The registry was established in June 1971 and includes CABG operations from March 1968 to October 1994. Data for the registry were obtained from five sources: (1) surgical data entry forms, (2) cardiac catheterization data entry forms, (3) computerized hospital records, (4) annual follow-up questionnaires mailed to patients and their families, and (5) medical record abstraction. The results of data analyis from the registry has been described in more than 100 articles published in prestigious clinical and epidemiological journals.

The patients evaluated in the present study were in the care of 3 cardiologists who have had a unique relationship with the registry. These cardiologists were practicing at St. Luke’s Medical Center before the development of CABG operation, and completed forms on virtually all of the patients that they catheterized. The operations were performed primarily by 11 surgeons; 9 of these had more than 100 patients in this data set, and 4 had more than 500 patients each.

The data set included information on the surgical procedures (reoperation, valve operation, number of grafts, number of sequential grafts, and use of internal mammary arteries) and patient risk factors (gender, acute myocardial infarction that occurred shortly before operation, hypertension, diabetes, cholesterol level, and history of smoking). Other patient risk factors were derived from the data sets. Patients were considered to have urgent operations if CABG operation and cardiac catheterization occurred on the same day. Left main coronary artery disease was defined as stenosis that obstructed at least 50% of the cross-sectional area of the left main coronary artery. The method of Rowe and associates [17] was used to combine the degree of coronary artery stenosis in the main coronary arteries and their branches, to give a total occlusion score that ranged between 0% for no occlusion to 300% for total occlusion of all three main arteries. Left ventricular function was measured using the wall motion data from ventriculography [18]. Wall motion in the left ventricle was assessed in six segments. Each of these segments was graded on a five-point scale from normal to dyskinetic. Analysis patients with any dyskinetic motion in two adjacent segments or abnormal wall motion in more than two of the six cardiac segments were considered to have compromised cardiac function. More complex classifications of left ventricular wall motion made virtually no improvement in the association between wall motion abnormalities and 30-day mortality rate. Ejection fraction was not used as a measure of ventricular function because it was often not available, particularly for patients during the first several years of the registry.

Demographic and clinical data of the patients are shown in three time periods: from 1968 through 1971, from 1972 through 1983 (the year before the introduction of percutaneous transluminal coronary angioplasty), and from 1984 through 1994. To more precisely track changes in patient risk and the introduction of new techniques, data were plotted each year from 1968 to 1992. After 1987, the annual number of CABG patients for study physicians declined because of increasing numbers of coronary artery disease patients sent to angioplasty, and the retirement of one of the cardiologists. As a result of small sample sizes in individual years after 1987, we created single data points on the grafts for the data from 1988 to 1989 and the data from 1990 to 1994.

Twenty-year Kaplan Meier survival curves were determined using the patients who had CABG operations without valve operations during the second time period, from 1972 (when operative mortality rates had stabilized) to 1983.

A logistic regression equation [19] was derived to find the association between patient risk factors and mortality within 30 days after operation. Only data after 1972 were used to derive this equation, because before 1972 the mortality rate had not yet stabilized, and including the earlier data would have obscured the effect of risk factors whose prevalence changed over time. For the analysis of patient risk and other analyses, cardiac catheterization data were necessary. Therefore, the 15% of the patients who were catheterized at other institutions and referred for CABG operations were excluded from all analyses.

By substituting information on the patient risk factors into the logistic regression equation, we computed the 30-day mortality risk for each patient in this study, including patients who had operations before 1972. The average of the mortality probabilities for all patients who had operations during that year was then used as the predicted mortality rate or risk for a given year.

To test for trends in risk-adjusted mortality, we used the logistic regression analysis, with 30-day mortality as the outcome variable. Trends were evaluated by testing the statistical significance of year of operation, when it was added as a regressor variable with all statistically significant patient risk factors.


   Results
Top
 Abstract
 Introduction
 Material and methods
 Results
Comment
 References
 
The clinical and operative characteristics of the patients are shown in Table 1. The data are compared for three time intervals to show trends. There were highly significant increases over time in the percentage of patients who are older than age 70, female, and hypertensive. There were also more operations on patients who had a recent myocardial infarction, or had left main disease, urgent operations, reoperations, and procedures that included a valve operation. There was only a minimal increase in the percentage of patients with diabetes or left main coronary artery disease, and no trend in the percentage of patients with compromised cardiac function. Cholesterol levels substantially decreased over time.


View this table:
[in this window]
[in a new window]
  		Table 1. Clinical and Operative Characteristics

 
Table 2 shows the results of the logistic regression analysis performed to derive an equation for the probability of 30-day mortality based on patient risk factors. The factors found to be associated with a greater probability of mortality, in order of decreasing statistical significance, were poor cardiac function, valve operation, age, reoperation, urgent operations (ie, operations on the same day as catheterization), female gender, and myocardial infarction shortly before operation. The area under the receiver operating characteristic (ROC) curve for this equation was 0.81.


View this table:
[in this window]
[in a new window]
  		Table 2. Logistic Regression Analysis for Patient Characteristics Independently Associated With 30-Day Mortality Rate Following CABG Operation

 
For each year from 1968 to 1992, the observed and predicted 30-day patient mortality rates are shown in Figure 1. The observed mortality rates decreased sharply from 1968 to1972, remained constant for a few years, and then increased gradually after 1976. The predicted mortality rates remained constant from 1968 to 1976, and then increased steadily. Because of increases in the predicted mortality rate, the ratio of the observed to predicted 30-day mortality rate actually declined over time. This decline was statistically significant in a logistic regression analysis for all data (p < 0.001), and even after eliminating data before 1972 (p = 0.01) and before 1976 (p = 0.02).



View larger version (24K):
[in this window]
[in a new window]
  		Fig 1. Observed and predicted mortality rates.

 
Trends in the use of internal mammary arteries, and the use of four or more grafts are shown in Figure 2. The use of internal mammary arteries increased to 63% in 1973, and then gradually declined to only 25% in 1979 and 1980. After 1980, the use increased again until internal mammary arteries were used in more than 80% of the cases in the 1990s. The use of four or more grafts increased gradually from 0% in 1969 to nearly 48% in 1979. The rates in 1980 and 1981 were lower than in 1979 (p < 0.05), but they increased again after 1981 to a maximum of 60% in the 1990s.



View larger version (22K):
[in this window]
[in a new window]
  		Fig 2. Percentage of cases that used specified grafting techniques.

 
Long-term survival rates for patients who had operations between 1972 and 1984 are shown in Figure 3. Patients who underwent operations in earlier periods had much higher mortality rates, and patients who underwent operations after 1984 had shorter follow-up time. The graph compares patients with vein grafts and patients with at least one mammary artery. Only patients with first time CABG procedures and no valve operations were used in this comparison. Survival rates are substantially better for internal mammary artery patients (p < 0.0001), and the difference between the survival probabilities increases with time. At 1 year the difference in mortality rates is 1%, at 5 years 4%, and at 10 years 9%. At 15 years the differences were 13%, and at 20 years 11%, although these differences were based on many fewer subjects.



View larger version (24K):
[in this window]
[in a new window]
  		Fig 3. Survival curves for first time, coronary artery bypass graft only operation according to whether or not mammary artery grafts were used.

 
Figure 4 shows the average number of sequential grafts. These grafts began to be used in 1971, and were used much more frequently in 1972. Then their use declined until 1976. Since 1981 there has been a consistent increase in the number of sequential grafts.



View larger version (22K):
[in this window]
[in a new window]
  		Fig 4. Average number of sequential grafts per vein.

 

   Comment
Top
 Abstract
 Introduction
 Material and methods
 Results
 Comment
References
 
The data analyzed in this study were used to track changes in CABG operations over more than two decades in patients under the care of 3 cardiologists in a private practice. Several changes were apparent over this time period. The most dramatic was the decrease in short-term mortality rates during the first 3 years of the operations, which reflected the learning curve for the operative team. After this period mortality rates increased, but this increase was due to higher risk for patients in more recent time periods. Individual risk factors that became increasingly common in the more recent time periods were older age, female gender (possibly because of diminishing rates of under diagnosis of coronary artery disease in women), myocardial infarction shortly before operation, hypertension, urgent surgery, previous CABG procedure, and concomitant valve operation. There were no changes over time in left ventricular function, and levels of coronary artery occlusion and cholesterol actually decreased over time. After taking into account patient risk, the adjusted mortality rate declined even after 1976. However, the rate of improvement was small after 1972.

There were also changes in the frequency with which certain procedures were used over time. The use of four or more grafts generally increased after 1969 to the present, with decline only in 1979 and 1980. Both internal mammary arteries and sequential grafts were widely used by 1972, followed by a decline in use until 1980, and a steep increase in use from 1980 to the present. The mortality rate for patients who had internal mammary grafts was substantially lower than for vein graft only patients, and differences in mortality rate increased with duration of follow-up. The data were inadequate to determine whether the differences in survival rates were due to type of graft, related surgical process variables, or patient selection.

Previous studies that reported temporal changes in patient and surgical factors differed from each other and from the present study, with respect to time periods, data collected, hospital type, and/or whether or not the patients had isolated CABG operations, valve procedures, or reoperations. Results from the previous studies are inconsistent with respect to changes in mortality rates over time. Mortality rates sometimes decline [1–5], remain unchanged [6–10], decreased initially then increased again [11], or increased [12]. As discussed below, most studies show increases in certain patient risk factors, which probably accounts for any increases in mortality rate.

Some risk factors, similar to those examined above, have been found in other studies to become more common in more recent time periods. These risk factors include older age [3, 5, 7, 10, 12–15], female sex [3, 7, 10, 12, 13, 16], hypertension [7, 16], diabetes [6, 7, 12, 16], left main coronary artery disease [9–11, 13, 15], extent of coronary artery disease [12, 13, 16], acute myocardial infarction [12, 14], and lower ejection fraction [11, 14]. The present study, and some previous studies, did not find evidence of declining left ventricular function as measured by worsening ejection fraction [5, 10], cardiac index [2], or congestive heart failure [13]. When the combination of all risk factors was evaluated, however, we found there was a clear increase in patient risk over time.

The large increases in the number of grafts, and in the likelihood of using a left internal mammary artery (LIMA), has been found in previous studies [7, 9–11, 13, 15]. Using data aggregated over several years, these studies showed that the increase in the use of the LIMA was consistent. In our data set the percentage of patients with a LIMA increased from 7% in 1968 to 65% in 1973, then gradually declined to 25% in 1979 and 1980, before a steep consistent rise to 75% in the late 1980s. The early decline in the use of LIMA was due to concerns about its effect on wound healing, and whether the blood flow through the internal mammary was adequate [20–22]. The value of the LIMA was only proved with experience over many years.

In contrast to the use of LIMA, the increase in the number of vein grafts was generally consistent, although there was a decline in 1980 and 1981. The pattern in sequential vein grafts was similar to that for LIMA, with a sharp increase from 1971 to 1972, a decline from 1972 to 1976, and then an increase from 1976 to the 1990s. Part of this increase was due to the increased numbers of reoperations. The decrease over several years in the number of sequential grafts performed probably reflected concerns with the procedure that were subsequently resolved.

Most of the previous studies were from major academic institutions, evaluated changes in shorter time intervals than the present study, and used grouped data. St. Luke’s Medical Center represents the unusual situation of a community hospital with a good database, which was developed early in the history of CABG operation. Because of this, database trends could be examined over a long period without piecing together results from multiple sources. Although it is possible that the findings from this study do not apply to other hospitals, they may demonstrate trends in many large community hospitals. [23]


   Acknowledgments
 
This work was supported in part by a generous grant from the Ladish Company Foundation, John H. Ladish Trustee, and the St. Luke’s Medical Center Research Foundation. The data for the study was provided by Jacob Assa of Cardiovascular Computer Systems Ltd. Suzanne Bentler, MS provided technical assistance.

We appreciate the contributions of all of the coronary bypass surgeons affiliated with the St. Luke’s Medical Center Cardiovascular Data Registry. We dedicate this paper to the memory of two of these surgeons, Dr Robert Flemma and Dr Richard Shore.


   References
Top
 Abstract
 Introduction
 Material and methods
 Results
 Comment
 References
 

  1. Rahimtoola S.H., Grunkemeier G.L., Teply J.F., et al. Changes in coronary bypass surgery leading to improved survival. JAMA 1981;246:1912-1916.[Abstract/Free Full Text]
  2. Miller C., Stinson E., et al. Discriminant analysis of the changing risks of coronary artery operations 1971–1979. J Thorac Cardiovasc Surg 1983;85:197-213.[Abstract]
  3. Rosenfeld K., Luft H.S., Garnick D.W., McPhee S.J. Changes in patient characteristics and surgical outcomes for coronary artery bypass surgery, 1972–82. J Public Health 1987;77:498-500.
  4. Keon W.J., Menzies S.C. Coronary artery disease bypass grafting patient profile. Contemp Card 1992;Aug/Sept:19-24.
  5. Jones E., Carver M., Guyton R.A., Bone D.K., Hatcher C.R. Trends in treatment of coronary disease today. Ann Surgery 1993;197:728-737.
  6. Cosgrove D., Loop F., Sheldon W.C. Results of myocardial revascularization. Circulation 1982;65(Suppl II):II37-II43.
  7. Davis P.K., Parascandola S.A., Miller C.A., et al. Mortality of coronary artery bypass grafting before and after the advent of angioplasty. Ann Thorac Surg 1989;47:493-498.[Abstract]
  8. Loop F., Lytle B., Cosgrove D.M., et al. Reoperation for coronary atherosclerosis. Ann Surgery 1990;212:378-386.[Medline]
  9. Akins C. Reoperative coronary grafting. Ann Thorac Surg 1994;58:359-365.[Abstract]
  10. Harris W., Mock M.B., Orszulak T.A., Schaff H.V., Holmes D.R., Jr Use of coronary artery bypass surgical procedure and coronary angioplasty in treatment of coronary artery disease. Mayo Clin Proc 1996;71:927-935.[Abstract]
  11. Haraphongse M., Na-Ayudhya R.K., Teo K.K., et al. The changing clinical profile of coronary artery bypass graft patients, 1970–89. Can J Cardiol 1994;10:71-76.[Medline]
  12. Naunheim K., Fiore A., Wadley J.J., et al. The changing profile of the patient undergoing coronary artery bypass surgery. J Am Coll Cardiology 1988;11:494-498.[Abstract]
  13. Cosgrove D., Loop F., Lytle B.W., et al. Primary myocardial revascularization. J Thorac Cardiovasc Surg 1984;88:673-684.[Abstract]
  14. Arcidi J. Trends in invasive treatment of single-vessel and double-vessel coronary disease. J Thorac Cardiovas Surg 1988;95:773-781.[Abstract]
  15. Christakis G. The changing pattern of coronary artery bypass surgery. Circulation 1989;80(Suppl II):I151-I161.
  16. Loop F.D., Lytle B.W., Gill C.C., Golding L.A.R., Cosgrove D.M., Taylor P.C. Trends in selection and results of coronary artery reoperations. Ann Thorac Surg 1983;36:380-388.[Abstract]
  17. Rowe G.G., Thomsen J.H., Stenlund R.R., et al. A study of hemodynamics and coronary blood flow in man with coronary artery disease. Circulation 1969;39:139-148.[Abstract/Free Full Text]
  18. Manley J.C., King J.F., Zeft H.J., Johnson W.D. The ‘bad’ left ventricle. J Thorac Cardiovasc Surg 1976;72:841-848.[Abstract]
  19. SAS Institute Inc, SAS/STAT User’s Guide, Version 6, 4th ed, vol 2. Cary: SAS Institute Inc, 1989.
  20. Singh H., Flemma R.J., Tector A.J., Lepley D., Jr, Walker J.A. Direct myocardial revascularization. Arch of Surg 1973;107:699-703.
  21. Flemma R., Singh H.M., Tector A.J., Lepley D., Jr, Frazier B.G. Comparative hemodynamic properties of vein and mammary artery in coronary bypass operations. Ann Thorac Surg 1975;20:619-627.[Abstract]
  22. Frazier B.G., Flemma R.J., Singh H.M., Tector A.J., Lepley D., Jr Physiological differences between vein and mammary artery as a coronary bypass. Rev Surg 1975:375-377.
  23. Tector A., Davis L., Gabriel R., Gale H., Singh H., Flemma R. Experience with internal mammary artery grafts in 298 patients. Ann Thorac Surg 1976;22:515-519.[Abstract]
Accepted for publication September 7, 1999.




This article has been cited by other articles:


Home page
 Ann. Thorac. Surg.Home page
W. F. Northrup III, R. W. Emery, D. M. Nicoloff, T. J. Lillehei, A. R. Holter, and D. P. Blake
Opposite trends in coronary artery and valve surgery in a large multisurgeon practice, 1979-1999
Ann. Thorac. Surg., February 1, 2004; 77(2): 488 - 495.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Thorac. Surg.Home page
M. Gaudino, F. Glieca, F. Alessandrini, G. Nasso, C. Pragliola, N. Luciani, M. Morelli, and G. Possati
High risk coronary artery bypass patient: incidence, surgical strategies, and results
Ann. Thorac. Surg., February 1, 2004; 77(2): 574 - 579.
[Abstract] [Full Text] [PDF]

Home page
 J. Thorac. Cardiovasc. Surg.Home page
S. Al-Ruzzeh, S. George, M. Bustami, K. Nakamura, S. Khan, M. Yacoub, and M. Amrani
The early clinical and angiographic outcome of sequential coronary artery bypass grafting with the off-pump technique
J. Thorac. Cardiovasc. Surg., March 1, 2002; 123(3): 525 - 530.
[Abstract] [Full Text] [PDF]

Home page
 Eur. J. Cardiothorac. Surg.Home page
N. Sadeghi, S. Sadeghi, Z. A. Mood, and A. Karimi
Determinants of operative mortality following primary coronary artery bypass surgery
Eur. J. Cardiothorac. Surg., February 1, 2002; 21(2): 187 - 192.
[Abstract] [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 Hartz, A. J.
Right arrow Articles by Assa, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hartz, A. J.
Right arrow Articles by Assa, J.

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