| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiothoracic Surgery, Michael E. DeBakey Department of Surgery, Houston, Texas
b Division of Cardiology, Department of Medicine, Baylor College of Medicine, Houston, Texas
c The Ben Taub Hospital, Houston, Texas
d The Michael E. DeBakey Veterans Affairs Medical Center, Houston, Texas
e The Texas Heart Institute at St. Luke's Episcopal Hospital, Houston, Texas
Accepted for publication December 18, 2008.
* Address correspondence to Dr Bakaeen, Department of Cardiothoracic Surgery, Michael E. DeBakey Veterans Affairs Medical Center, OCL 112, 2002 Holcombe Boulevard, Houston, TX 77030 (Email: fbakaeen{at}bcm.edu).
Presented at the Fifty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Austin, TX, Nov 5–8, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Methods: Using prospectively collected data from our departmental database, we identified all primary, isolated CABG operations (n = 1,042) performed between July 1997 and April 2007. Operations were then stratified according to the seniority of the primary surgeon: first-year cardiothoracic resident (CT1), second-year cardiothoracic resident (CT2), or staff surgeon. Data were examined for any association between seniority and surgical outcomes.
Results: Staff, CT2, and CT1 surgeons performed 47 (4%), 610 (59%), and 385 (37%) cases, respectively. Efficiency was correlated with experience: for CT1, CT2, and staff surgeons, respectively, operative times averaged 345, 313, and 302 minutes; perfusion times averaged 118, 106, and 96 minutes; and cross-clamp times averaged 68, 58, and 57 minutes (p < 0.05 for all comparisons). The incidences of major morbidity (10.1%, 12.3%, 12.8%) and operative mortality (0.8%, 1.5%, 2.1%) were similar after operations performed by CT1, CT2, and staff surgeons, respectively (p > 0.15 for all). In univariate and multivariate analyses, the seniority of the primary surgeon did not independently predict morbidity or perioperative mortality. On follow-up (mean, 1,485 ± 1,015 days), there was no significant difference in patient survival (log-rank, p = 0.64).
Conclusions: Lower academic seniority was associated with longer CABG operative times but did not affect outcomes. Thus, training residents to perform CABG is safe and is characterized by progressive improvement in their technical efficiency.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
As heart surgeons, we are fully aware of the importance of quality assurance, and our specialty has led the way, by various professional associations and platforms, toward setting higher standards of care and accountability. Yet, the responsibility of training the next generation of cardiac surgeons may pose at least a theoretical risk of compromising outcomes. Academic centers are faced with the challenge of training residents and delegating responsibility without compromising patient safety. In the field of cardiothoracic surgery, hands-on experience is critical for trainees as they follow the steep learning curve this discipline requires.
Various academic centers have evaluated the safety of training residents to perform cardiac surgery [1–12]. However, most studies on this topic have not specifically looked into the effect of the residents' level of seniority on outcomes. Furthermore, these studies have typically addressed only immediate perioperative results and have lacked substantial follow-up. The purpose of the present study was to determine the impact of cardiothoracic trainees' graduate level on the short- and mid-term outcomes of isolated coronary artery bypass grafting (CABG) surgery at a single academic center where residents perform the vast majority of cardiac cases.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Academic Setup
The Michael E. DeBakey Veterans Affairs Medical Center is academically affiliated with Baylor College of Medicine (BCM) and serves as one of its primary teaching hospitals. At the time of the study, the BCM cardiothoracic surgery residency program was a 2-year program, admission to which required successful completion of at least a 5-year general surgery residency. Throughout the academic year, 2 cardiothoracic residents at a time rotate at the MEDVAMC. A first- and a second-year resident work together in 4-month blocks, starting on the first day of July. The cardiac surgery residents are important members of the team and actively participate in all stages of patient care delivery. In the operating room (OR), the resident usually acts as the primary surgeon under the close supervision of faculty members.
Conduct of Surgery
During each operation, the resident harvested the internal mammary artery, and the staff surgeon or a physician assistant simultaneously harvested the saphenous vein. For the construction of the distal anastomoses, the heart was arrested with cold blood potassium cardioplegia administered through antegrade and retrograde routes for the duration of the aortic cross-clamping. A partial cross-clamping technique was used to construct the proximal anastomoses. The resident operated from the right side of the table, while the staff surgeon supervised from the left side. The degree of staff participation in surgery typically depended on the skill and competence of the resident, as well as on case-related factors. The general policy at the MEDVAMC is to allow the resident to be the primary surgeon and perform the critical steps of the CABG, including at least 50% of the distal anastomoses. On rare occasions, including some emergency or high-risk cases, a staff surgeon was the primary surgeon, either because no resident was available to perform the case or because it was the very beginning of the resident's rotation at the MEDVAMC and the resident did not yet attain the necessary skill level. The attending staff usually decides before the beginning of the operation on who will be the primary surgeon and that decision may change during the operation depending on the condition of the patient and the conduct of the surgery.
Outcomes
Three outcome variables were evaluated separately in this study: perioperative morbidity, 30-day operative mortality, and patient survival. The CICSP provides semiannual risk-adjusted outcomes reports to contributing centers. In CICSP, 30-day operative mortality is defined as the number of deaths that occur during the index hospitalization or within 30 days after surgery, plus any deaths that occur more than 30 days after surgery that are the direct result of a perioperative surgical complication. The expected mortality for each patient is calculated using the CICSP risk model which is time-sensitive and is based on the 3 years of VA national data that immediately preceded the release of the semiannual report. Perioperative morbidity is defined as the presence of any of the following major complications, alone or in combination: endocarditis, renal failure necessitating dialysis, mediastinitis, reoperation for bleeding, being kept on a ventilator for longer than 48 hours, repeat cardiopulmonary bypass or mechanical circulatory support to treat a complication, stroke, and coma for longer than 24 hours. This complication assessment uses standard definitions and criteria adopted by the VA Cardiac Surgery Consultants Board. For all patients, survival information (through June 2007) was obtained from the Social Security Death Index and CICSP.
Statistical Analyses
Data are presented as mean ± standard deviation unless otherwise specified. Baseline and demographic characteristics were compared between groups by using analysis of variance for continuous variables and the 
2 test for categoric variables. Nonparametric methods were used for variables that were not normally distributed. Stepwise, multivariable logistic regression modeling was used to identify independent risk factors for perioperative morbidity and mortality; a significance level of 0.05 was used for both entry and selection. Similarly, multivariable Cox proportional hazards modeling was used to identify independent risk factors for all-cause mortality. Covariates under consideration for all models were all baseline characteristics (as listed in Table 1), surgical priority, and the operator level of the primary surgeon (ie, CT1 resident, CT2 resident, or staff surgeon). Overall survival was estimated by using the Kaplan-Meier method and was expressed as a percentage ± 95% confidence interval (CI). Analyses were performed using STATA 9.0 (STATACorp, College Station, TX).
|
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Data on surgical procedures are summarized in Table 2. Proportionally, staff surgeons performed more urgent and emergent cases. There were no intergroup differences in the mean number of bypassed vessels or the use of internal mammary arterial grafts. The documented mammary artery injury rates were 0.27%, 0.69%, and 0% by CT1, CT2, and staff, respectively. In contrast, total OR time, operating time, bypass time, and ischemia time decreased with increasing operator seniority.
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
In our study of predominantly male veterans undergoing isolated primary CABG surgery, we found that the level of experience of a supervised resident was not associated with perioperative mortality or morbidity. Our findings are consistent with others reported in the cardiothoracic surgical literature [1–12]. Guo and colleagues [5] examined perioperative outcomes among 2,906 isolated CABG operations and found no differences between trainees' and staff surgeons' patients in perioperative mortality, composite morbidity, length of intensive care unit stay, or length of hospital stay; however, the authors did not account for baseline differences in their analyses. Baskett and colleagues [1] performed multivariable analyses of data from 3,004 isolated CABG procedures and found that operations performed by residents did not have poorer perioperative outcomes than operations performed by staff surgeons. Also, in off-pump CABG series, operations performed by trainees appear to have similar outcomes to those performed by staff surgeons [6–10], even in high-risk patients [10]. The same appears to be true in operations for valvular heart disease; multivariable analyses have shown that these procedures produce no more adverse outcomes when performed by trainees than when performed by staff surgeons [1, 11, 12].
In this study, we went further by evaluating the impact of the residents' stage of training and also by evaluating midterm outcomes. Trainee level has not been shown to be associated with adverse outcomes of operations performed in other surgical subspecialties, such as operations for strabismus [15], biliary procedures [16], and laparoscopic appendectomy [17], although some findings suggest that there may be a higher recurrence rate after open inguinal hernia repairs performed by junior residents [18] and more operative complications in laparoscopic cholecystectomies performed by junior residents [19].
There are at least two reasons why the outcomes of operations performed by trainees may be similar to those performed by staff surgeons. First, as our study and other published series have shown, cardiothoracic residents usually operate on lower-risk patients [5–7, 10]. This tendency has been ascribed to the selection of appropriate-risk patients by staff surgeons so as not to compromise patient outcomes [20]. Indeed, in the present study, cases performed by staff surgeons had a higher degree of acuity and included a greater proportion of patients who required preoperative IABP support reflected in a greater expected operative mortality rate than cases performed by residents.
Second, when residents perform procedures, staff surgeons are available inside and outside the OR to provide the necessary supervision. One study has associated this availability with reduced rates of adverse perioperative outcomes [21], although other studies have not corroborated this finding [10, 20].
Cardiac trainees' operative and perfusion times are inversely proportional to the trainees' level of surgical experience [1, 17–19]. Therefore, it is not surprising that, despite the similarities in the number of bypass grafts performed per case, the operative and perfusion times were longest for CT1 residents and shortest for staff surgeons. However, these longer operative times did not translate into adverse outcomes or prolonged hospital stays.
Hospitals are under relentless pressure from the health care marketplace to publicly report surgical outcomes. Pay for performance has already been incorporated into health care finances and is here to stay, and the medico-legal environment is often counterproductive. The development of cardiac surgical databases, including the Society of Thoracic Surgeons database and the VA CICSP, in addition to various practice guidelines, affirms the commitment of cardiac surgeons to accountability and superior standards of care.
Our data and those of other authors seem to show that the care of patients is not compromised by the process of training surgical residents. This has been shown in a wide range of cardiothoracic patient populations and operations with varying degrees of surgical complexity. This result appears to be consistent across different cardiothoracic training systems, including the Canadian [1, 2], British [6–10], and American systems. However, the evolution of new technologies, including minimally invasive and robotic techniques, is likely to challenge surgeon educators as they struggle to master such techniques themselves before being able to teach them effectively and safely in the current modus operandi of our cardiothoracic training system. The concept of virtual-reality surgical training is rapidly evolving and provides an opportunity for a new training paradigm for all surgeons at different stages of their career [22, 23]. The novice surgeon will need to prove a minimum and safe level of technical proficiency in the laboratory before the privilege of operating on patients. Educational funds should be set aside for supporting the establishment of such simulation laboratories at major cardiac training facilities.
Our study has limitations similar to those of other studies in this area. It involved a single center, it was retrospective (although data were collected prospectively), and almost all of the patients were men. Also, the study spanned a considerable time period, during which surgeons and residents came and went, and preferences for specific surgical techniques and medical therapies changed. The surgical volume and case acuity increased at our institution in recent years resulting in disproportionately more cases performed by staff toward the end of the study period with shorter follow-up available for their patients. The latter, coupled with small number of operations performed by staff surgeons, may have reduced the power of our analyses to detect significant differences between trainees and staff surgeons. Furthermore, although multivariable analyses were used to determine the association of graduate level with outcomes, residual confounding unmeasured covariates may still have been present. For example, we did not have data on transfusion requirements, freedom from angina, and coronary reintervention. Finally, perioperative mortality was uncommon (13 out of 1,042), so analyses for this endpoint should be interpreted cautiously.
In summary, we found no association between residents' level of experience and CABG outcomes. Lower academic seniority was associated with longer operative and perfusion times, but these did not translate into adverse effects. Thus, academic centers can train residents in CABG surgery without compromising patient safety, as long as staff members provide appropriate supervision and guidance to offset the inexperience of new or inexperienced residents.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Like you and your colleagues, I have always believed that hands-on experience is critical for the proper development of our cardiothoracic residents, and I know a lot of people in the audience feel similarly. I have always been a believer in Dr Shumway's adage, which I am paraphrasing here, which is, "The hardest thing about heart surgery is just getting to do it." It will be good to have the information in your study in the literature. We have mentioned this need to document the validity of our practices, even in the realm of education, several times at this meeting already, especially as we need to be more and more transparent. We have got to be ready to talk to our patients, our payors, and our hospital administrators, about the fact that, as we train surgeons, we can maintain the safety of our patients as well as efficiency in the delivery of their care.
It is a struggle for us, those of us who are teaching young surgeons how to operate, to figure out not only what would we be willing to let people do but how much to let them do at any given time. I believe that the answer to that question is that our decisions need to be about the increment. I believe that we expect to teach and learn all the time. Our patients, if we ask them, want us to do this also. They just want us to assure them, that we don't do too much, that we don't advance too much, in what we are teaching on any one patient or at any one time. So I think, if we really think through this question, that the answer to my rhetorical question really is that it is all about the increment, whether it is a new procedure we are doing or a new resident we are training.
I thought it might be nice to hear you talk a little bit about how your residents decide how they categorize a case, with regard to whether or not they can consider themselves the primary surgeon. This is a hot topic on resident Web sites and blogs these days. In fact, there is fair amount of relatively bitter discussion about how the attendings count cases for the residents that the residents don't think they have really done. I also wonder whether the approach to resident involvement is different at different hospitals in your system. In other words, is this commitment to having the residents do cases more common at the VA and less common at other hospitals? Thanks again for the privilege of discussing this paper.
DR TSAI: Thank you, Dr Tribble, for the kind comments and important questions. At Baylor College of Medicine, we are very focused and committed to cardiothoracic training and providing a comprehensive experience for the residents. At the VA, and at the Ben Taub General Hospital, residents stand on the right side, the surgeon's side, of the operating table. The staff surgeon guides the resident through the various steps of the procedure including cannulation, cross-clamping, and also in terms of the distal anastomosis. I can remember when I was training I was given the opportunity to perform a distal anastomosis of the LIMA to LAD [left internal mammary artery to left anterior descending coronary artery] at the very beginning of my rotation. However, as you pointed out, the key is to gradually and safely delegate surgical steps to the residents depending on their level of surgical skill and ability. The pace of progression from an assistant surgeon to a primary surgeon can vary from one resident to another and depends on the nature and complexity of the case. For CABG operations, a resident is regarded as the primary surgeon if he or she performs the critical portions of the operation, including at least 50% of the distal anastomoses. In our experience, for a straightforward CABG [coronary artery bypass grafting], this can be accomplished quite early in the training process. With good mentoring, the residents rapidly develop a very positive attitude and feeling about the training experience. We hold dearly to the principles that have been set by pioneers in our speciality, such as Dr Shumway, that emphasize good mentorship and hands-on experience. We have the responsibility to take good care of our patients and to train our residents who will be the backbone of a successful and competent future workforce.
DR FREDERICK GROVER (Denver, CO): I just want to comment, that was a very nice paper, well done, great analysis, and it really stresses the importance, I think, of even though their cases take longer, if you are patient and you really give good supervision and you exercise some degree of stratification of what a resident's ability is in terms of what cases they do, you can deliver very, very good care. And I would suspect if you risk adjusted all of those, the risk-adjusted mortality would probably be just about equal right across the three groups. And I think it is really important to have this in the literature. It reinforces the ethics of our teaching programs and the morality of it to the general public when it appears, and I congratulate you on a nice piece of work.
DR TSAI: Thank you Dr Grover. We agree with your comments and our risk-adjusted outcomes support that notion.
DR ARA VAPORCIYAN (Houston, TX): Great work, and it is an important topic. One question, I think what your paper does highlight is the lack of a good, reliable, and valid measurement of surgical skills. Using outcome measures is something that has been proposed to do this, but I think the underlying theme in your work has to be that not only can we train them safely but are we training them? And so the question I would ask you is, since the outcome measures clearly don't distinguish between a staff surgeon doing a case and a first-year resident doing a case, do you have any other data that shows that they actually are learning the skill and they are improving as they go along in their training?
DR TSAI: Thank you, Dr Vaporciyan, for that interesting question. I think in our training institution, with the commitment from the teaching standpoint, it is critical to look at what kind of skill-set a resident brings forth from whatever general surgery program he or she have completed. It is our commitment to work with them from that baseline level of technical ability and gradually and efficiently build it up. It is the commitment of spending that time, pushing that resident to attain the fullest capability and the ability to operate. And as you go through the cases within a four-month span, it is amazing how you can see these residents mature with a competent surgical skill set. The good surgical outcomes are reassuring but the real measure of skills learned is tracked by the performance level of the resident in the operating room.
DR WILLIAM A. GAY, JR (St. Louis, MO): I think this is a very important piece of work and I am anxious to see it in print because it is important for it to be in print. As everyone in this room realizes, patient safety is right up there at the forefront of what the media, what the payers of health care, and what we should be all about. That having been said, there are a number of variables that affect outcomes, and you have listed a great many of them. I am assuming that there was one attending surgeon, maybe not. If not, that is a variable, because everybody in this room does things a little bit different than their colleagues and maybe is a little bit more liberal in what they allow or what they want the resident to do and their method of teaching. So I think you might go back and look again at the differences in the policies and the practices of the attendings as another variable.
But this is a good study and I think it is one that desperately needed to be done. Thank you.
DR TSAI: Thank you for your comment. All the attendings involved during the study period were trained at the same institution and generally adhered to the same training paradigm. That being said, we agree that different staff surgeons may have had different approaches for a certain set of circumstances and this is potentially one of the limitations of the study.
DR JOHN H. CALHOON (San Antonio, TX): I enjoyed your presentation. I have just one comment. I would echo all the other comments about how important it is to be in the literature, but it struck me from a maintenance of certification standpoint, how does the faculty maintain his own competence at taking down a mammary when the faculty is a saphenologist and not the mammary artery procurement person and is that a concern of yours? Where do you keep current on doing the procedures when you did 47 CABGs over a 10-year period or you and two or three faculty, as Dr Gay pointed out? I wish you could address that for me. Thank you.
DR TSAI: Thank you, Dr Calhoon. Those are good questions. As faculty, again, being the assistant from the teaching side is actually a more difficult job I think in my mind than doing the operation, because you have to predict every step, you have to be a second ahead of the resident, and once there are some complications with the stitches, for example, of an anastomosis, if they cannot fix it, you are basically there to repair it. I think we have a selective group of faculty where we are actually more into education, we push the limit, and we are actually comfortable with the operations ourselves. A LIMA takedown, for example, if the residents are honest with themselves and they have difficulty in taking them down, particularly, they will take down about 90% of the LIMA, for example, and if they are not confident in that last centimeter because they are worried that they are going to injure the LIMA itself, they will actually call us in and we will be expected to take it down.
In terms of practices, I think overall the fact that you are assisting, that kind of experience will probably be enough. There are also other cases, for example, the cases of the 47 which are emergent and urgent, typically the scenario is the residents are involved in other cases and they are physically not available for those cases, and that is when the faculty will step in and do the cases themselves.
This paper only addresses the CABGs we do. The faculty will have more participation in redo cases, valve cases, and other aortic operations. Those cases will supplement and tune-up of our own surgical skills through time.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. G. Bakaeen, E. H. Stephens, D. Chu, W. L. Holman, A. A. Vaporciyan, W. H. Merrill, and F. L. Grover Perceptions regarding cardiothoracic surgical training at Veterans Affairs hospitals J. Thorac. Cardiovasc. Surg., May 1, 2011; 141(5): 1107 - 1113. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Martin, M. J. Eckert, W. E. Eggebroten, and A. C. Beekley A New and Simplified Technique for Laparoscopic Gastric Bypass in a Residency Training Program: Decreased Resource Utilization and Enhanced Training Arch Surg, September 1, 2010; 145(9): 844 - 851. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. T. Hosseini, I. Saeed, K. Mandal, A. Kourliouros, O. Valencia, and M. Jahangiri Short-term outcome following cardiac surgery - a comparison between consultant and trainees' performance Interact CardioVasc Thorac Surg, June 1, 2010; 10(6): 889 - 891. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. G. Bakaeen, J. Huh, S. A. LeMaire, J. S. Coselli, S. Sansgiry, P. V. Atluri, and D. Chu The July Effect: Impact of the Beginning of the Academic Cycle on Cardiac Surgical Outcomes in a Cohort of 70,616 Patients Ann. Thorac. Surg., July 1, 2009; 88(1): 70 - 75. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
