Changing Patterns of Resident Operative Experience From 1990 to 1997

Hypothesis  Resident operative experience has increased or decreased with respect to 12 specific operations.

Design  A retrospective analysis of resident operative experience reported to the Accreditation Committee for Graduate Medical Education for academic years 1990-1997.

Subjects  Residents completing an Accreditation Committee for Graduate Medical Education surgical program.

Main Outcome Measures  The total number of residents, average number of operations performed per resident, and the most common operations performed.

Results  The number of house staff completing surgical residency training programs has remained constant, while operative volume has increased from 1991 to 1997. Comparison of the frequencies of 12 selected operative procedures performed in academic years 1990-1991 and 1996-1997 found increases in the following procedures: carotid endarterectomy (137%), pancreaticoduodenectomy (66.7%), laparoscopic cholecystectomy (64.8%), parathyroidectomy (51.2%), thyroidectomy (19.2%), colectomy (14.1% to 44.4% depending on subtype), and elective infrarenal aortic aneurysm repair (10.7%). Conversely, frequencies decreased for open cholecystectomy (63.4%), open parietal cell vagotomy (40%), modified radical mastectomy (15.2%), gastroesophageal antireflux procedure (10.4%), and subtotal gastric resection (8.93%). Resident experience was essentially unchanged for emergent infrarenal aortic aneurysm repair and laparoscopic proximal gastric vagotomy.

Conclusions  The number and variety of operative interventions in surgical therapeutics is changing. Continued analysis of the operative experience during surgical training will indicate the need for changing requirements for surgical resident experience. The causes of these shifts are not specifically addressed by this study. Perhaps technological advances in the diagnosis and management of surgical patients or the increase in subspecialty training programs have affected the experience of general surgery trainees.

THE AMERICAN health care system has been in transition for the last 2 decades. Principles of cost containment and efficiency have pervaded our hospitals.¹ Technological advances with significant inherent cost continue to affect modern medical and surgical management. Advances such as duplex scanning, high-resolution computed tomography, ultrasonography, magnetic resonance imaging, and endoscopy combined with the advent of minimally invasive surgery have changed the diagnostic and therapeutic practices of surgery in this country.² These changes have affected the characteristics of surgical training as well. The number of surgical procedures has increased,^3,4 while the variety of operative experiences during surgical training has changed over the last decade. Awareness of these changes may assist program directors to better structure house staff operative experiences and requirements.

The Residency Review Committee (RRC) for Surgery, under the direction of the Accreditation Council for Graduate Medical Education, annually compiles data on all recorded procedures performed by surgical residents in accredited surgical training programs in the United States. A computerized national database was initiated by the RRC in 1986. At the conclusion of surgical training, chief residents are required to submit their operative experience, confirmed by the program director, to the RRC and the American Board of Surgery in the form of a surgical operative log. The surgical operative logs are compiled by the RRC and the data are printed in annual reports called resident statistic summaries.

The RRC published results for academic years 1990-1991 to 1996-1997 were analyzed for the following 11 arbitrarily selected operations: pancreaticoduodenectomy, colectomy, open and laparoscopic cholecystectomy, elective and emerging infrarenal aortic aneurysm repair, carotid endarterectomy, thyroidectomy, parathyroidectomy, modified radical mastectomy, gastroesophageal antireflux procedure, open and laparoscopic parietal cell vagotomy, and subtotal gastric resection.

The data consisted of numbers of residents and overall and facility statistics summarizing the frequencies of specific operative procedures: mean, mode, maximum, and SD. The main outcome measurements were average number of operations performed per resident and the most common operations performed. The total number of procedures was divided by the total number of residents to obtain the average number of procedures performed per resident. The changes in operative volume from academic years 1990-1991 to 1996-1997 were expressed as a percentage increase or decrease each year. Analysis of variance was used to evaluate the statistical significance of changes in the average numbers of operations. The χ² test was used to compare categorical variables. Significance was recognized as P<.05. All statistical analyses were made using Statistical Product and Service Solutions for Windows, version 6.1 (SPSS Inc, Chicago, Ill).

The total number of major procedures performed by surgical house staff has increased from 914,812 cases to 976,613 cases (6.8%), while the major case categories have remained unchanged in this period. The number of residents completing surgical training has remained relatively unchanged (1015 graduates in 1991 and 1016 graduates in 1997).

Comparison of the 12 selected operations performed in academic years 1990-1991 and 1996-1997 found statistically significant increases in carotid endarterectomies (137% increase, P<.001), pancreaticoduodenectomies (66.7% increase, P<.001), laparoscopic cholecystectomies (64.8% increase, P<.001), parathyroidectomies (51.2% increase, P<.001), thyroidectomies (19.2% increase, P<.001), colectomies (14.1% to 44.4% increase depending on subtype, P<.006), and elective infrarenal aortic aneurysm repairs (10.7%, P<.002) (Table 1). The operative volume of colectomies has increased across all subtypes. Partial colectomies have increased by 14.1% (P<.001). Ileoanal pull-through subtype exhibited the largest increase (44.44%, P=.002) (Table 1). Interestingly, there was a major decrease in the number of ileoanal pull-through procedures in the 1996-1997 academic year despite an overall increase in academic years 1991-1997.

Resident experience of the following procedures decreased: open cholecystectomies (63.4% decrease, P<.001), open parietal cell vagotomies (40% decrease, P<.001), modified radical mastectomies (15.2% decrease, P<.001), gastroesophageal antireflux procedures (10.4% decrease, P=.02), and subtotal gastric resections (8.93% decrease, P=.05) (Table 1). There were no statistically significant changes in the resident experience of emerging infrarenal aortic aneurysm repairs or laparoscopic proximal gastric vagotomies.

Comparison of the reported house staff operative experience in 1991 vs 1997 finds that the number and variety of surgical procedures performed by surgical residents in this country is changing (Table 2). Overall volume is increasing while certain procedures seem to be decreasing at teaching hospitals. These changes can be attributed to technological advances and the resultant evolution of surgical practice rather than any change in the diseases of our patient population, although this cause cannot be excluded with certainty. Whether these shifts represent the nationwide surgical experience of board-certified surgeons requires more extensive analysis, including examination of data describing operations performed in nonteaching and private hospitals.

The average number of total cholecystectomies (open and laparoscopic) performed per resident increased from 59 cases in 1991 to 90.8 cases in 1997. The RRC began to classify laparoscopic cholecystectomy as a separate procedure from the encompassing cholecystectomy category in 1994. The separation of these 2 procedures in 1994 exhibited a precipitous decrease in the number of open cholecystectomy procedures from 68 operations per resident in 1993 to 21.6 operations per resident in 1997. If this decline continues, the issue of adequate experience with open cholecystectomy procedures may affect the experience of surgical graduates.⁵ The overall increase in cholecystectomy from 1991 to 1993 (12.96%) may represent both more accurate diagnostic tests and an increased adherence to surgical recommendations for patients with cholelithiasis who objected to an open cholecystectomy procedure. Laparoscopic cholecystectomy has become a fundamental procedure (gold standard) at teaching and nonteaching hospitals. Recent data suggest that, while operative duration is increased, the conversion rate, the incidence of complications, and the postoperative stay are identical in teaching and nonteaching surgical practice.⁶

The use of videoscopic surgery has provided surgeons an opportunity to approach surgical pathology in many anatomical locations in an alternative manner.⁷ The increased prevalence of endoscopic procedures has prompted the RRC (surgery) to establish a separate category for all endoscopic procedures in the resident statistic summaries. However, not all endoscopic procedures have increased in popularity. For instance, the volume of both open and laparoscopic parietal cell vagotomies has decreased (40%, P<.001 and 55.71%, P=.37, respectively). We cannot, however, account for the dramatic increase in laparoscopic parietal cell vagotomies in the 1996-1997 academic year (34.78%) because the RRC database reveals a weak downward trend since 1985. Continued evaluation of 1998 and 1999 RRC data is necessary to see if this increase is an aberrance or a continued increase in the use of laparoscopic parietal cell vagotomy.

The most notable increase in our study was the almost exponential increase in carotid endarterectomies. The increase has propelled the procedure into 1 of the 10 most frequent procedures performed in our teaching hospitals and the second most common vascular procedure performed by general surgery residents, preceded only by bypass procedures (Table 2). The reason for this increase may be the decreased morbidity and mortality associated with the procedure, contributing to more aggressive surgical management as the result of the NASCET trials that supported surgical intervention in critically stenotic yet asymptomatic patients. Cebul et al⁸ reported that 47.5% of carotid endarterectomies performed in Ohio were in an asymptomatic patient population or patients exhibiting nonspecific symptoms.

Experience with thyroidectomy and parathyroidectomy procedures among house staff is increasing. The variability of experience across programs has been noted in prior studies. Of a reported 268 surgical training programs, 63.8% have previously reported inadequate experience in parathyroid surgery.⁹ The average number of procedures increased from 12 to 14.3 procedures for thyroidectomy and 4.1 to 6.2 procedures for parathyroidectomy, implying that despite this current increasing trend, many residents continue to report the minimum requirements for certification.

The overall increase in surgical volume is encouraging for surgical education. Although there had been only a 2.1% increase in the number of general surgery graduates since 1987,¹⁰ the increase in the number of operative procedures performed in teaching programs indicates that training has improved. It has been reported that operative experience usually results in increased proficiency and, ultimately, decreased patient morbidity and mortality.^11-14 As indicated in Table 2, the variety of operative procedures remains relatively constant. However, as shown in this study, certain procedures gain or lose utilization in teaching institutions. One cause seems to be the subspecialty introduced in a particular program. Galandiuk¹⁵ demonstrated that there was a significant increase in the number of colorectal procedures performed at a university hospital after the acquisition of a colorectal surgical subspecialist. In a generation where a larger number of US graduates are seeking subspecialty training than 2 decades ago,¹⁰ the ratio of specialists to general surgeons will increase and possibly make specialized procedures commonplace among future surgical trainees.

The implications of the changes reported by this article are difficult to predict. Future analysis of trends in procedures other than the 12 reported here could exhibit similar patterns. One could speculate that current trends indicate a propensity toward minimally invasive surgery. As the learning curves of surgeons with regard to videoscopic surgery increase, these trends will undoubtedly be reflected in training programs. An inventory of the increase in centers of excellence for laparoscopic surgery would likely support this claim. The trend toward less-invasive procedures is mirrored in oncologic surgery as well. Report C will probably begin to distinguish axillary sentinel node biopsies as a separate operative category. It will be interesting to follow house staff experience with axillary node dissections compared with sentinel node biopsies. It would not be surprising if current data indicated a decline in the number of axillary node dissections performed in our teaching institutions. In addition, the sentinel node technique is becoming the standard of care in the diagnostic evaluation of melanomas. Technology is permitting a new approach to operative management of our patients. The outcome will undoubtedly lead to decreased experience and proficiency with "open" techniques. The hypothetical question becomes "How much experience is too little?"

The changes reported here necessitate periodic analysis by program directors, the RRC, and the American Board of Surgery for continued vigilance to the evolution of general surgical training in the United States. Additionally, possible adjustments may be needed in the minimal requirements in the primary components of general surgery for credentialing. The effect that this will have on future surgical trainees is difficult to predict and requires continued surveillance.

Presented at the 107th Scientific Session of the Western Surgical Association, Santa Fe, NM, November 17, 1999.

Corresponding author: Claude H. Organ, Jr, MD, Department of Surgery, University of California, Davis, East Bay Program, 1411 E 31st St, Oakland, CA 94602 (e-mail: chorgan@ucdavis.edu).

J. Bradley Aust, MD, San Antonio, Tex: I am sorry Dr Organ isn't here to hear the fine presentation that Dr Parsa made. My comments relate more philosophically than directly to the numbers that are presented. The operations I did as a resident were not those of today's residents; subtotal gastrectomy for duodenal ulcer, radical mastectomy for cancer of the breast, varicose vein surgery, stripping for varicose veins, subtotal thyroidectomy for Graves disease, hemipelvectomy for sarcomas of the lower extremity, and lung resections for tuberculosis. During my residency, vascular surgery started, abdominal aortic aneurysms were first resected, bypass vascular surgery and carotid endarterectomy were developed. Open heart surgery began, as did transplantation of kidneys. In the ensuing years medical treatment took over the treatment for duodenal ulcer, limited surgery sufficed for breast cancer, laparoscopic surgery replaced much of traditional open surgery.

I would like to emphasize most fervently that surgeons should be taught to do surgery, not operations. How to cut, dissect, suture, how to deal with the various tissues, the gut, the liver, the pancreas, and how to use the various tools that have been developed besides the knife and the scissors, to include the Bovie, which general surgeons adopted during my residency, staplers, and, more recently, endoscopes and laparoscopes. The key to surgery is judgment, namely, whom to operate on, what to do, and when to do it. I decry the use of the operative checklist to decide the scope of surgical practice of a given surgeon. The scope of general surgery is so broad that no surgeon can do one of every listed operation every year and some operations once every 5 years.

The RRC is charged with the responsibility for setting the minimum standards and wants assurance that prospective surgeons have had enough operative experience. This requires the collection of operative numbers to ensure sufficient volume of cases of a broad variety to encompass the different technical requirements for operating on the liver, the spleen, the pancreas, the gut, the lung, the alimentary tract, and the blood vessels. In addition, they need exposure to an array of difficult and complicated surgical experiences. That the numbers of our current residents are increasing pleases me, especially when the residents have more opportunities to do the Whipple operation and other complicated surgical procedures. There is no question that this generation of surgeons will become laparoscopic surgeons. My hope is that they will be sufficiently trained to satisfactorily convert the failed laparoscopy cases to open surgical procedures.

Bruce L. Gewertz, MD, Chicago, Ill: One of the persistent controversies in surgical education has been the development of fellowships, and one argument is that fellowships detract from the number and complexity of cases done by the general surgery resident. The opposing argument, advanced by card-carrying vascular surgeons such as myself, is that the establishment of centers of excellence in vascular surgery will increase the number of complex cases coming to the medical centers where our residents are training, and actually increase the numbers of cases done by general surgery residents.

I was very intrigued by the data here that suggest that because of the greater enthusiasm for operating on asymptomatic carotid lesions and the earlier detection of aortic aneurysms, our general surgery residents are doing substantial amounts of vascular surgery during their training. This seems to fortify the argument that vascular fellowships are beneficial for the training of general surgery residents. I wondered if the database will allow you to pull out the residencies in which fellows and residents are in the same program to see if vascular surgery training programs have a salutary or detrimental effect.

Richard A. Prinz, MD, Chicago: I would like to know if you know the total number of operations performed by each resident reporting to the RRC. I am interested in this because I wonder if the average number of cases each resident is reporting now is greater than in the past. I would also like for you to editorialize on this to see if you think that our residents may be spending too much time in the operating room.

I would like to ask about operations that were done when I was a resident such as carotid endarterectomy, thyroidectomy, and parathyroidectomy. Although there have been some technological changes that affect these particular operations, I don't think that they could explain the increase in the performance of these procedures. So I would like to ask if you think the increase that you are reporting in these operations reflects the subspecialization that is going on in surgery. In other words, does this support the idea of having an expert in an area in your training program who will bring more operative experience to your residents?

Jerry M. Shuck, MD, Cleveland, Ohio: In evaluating a resident's experience, both evaluating groups, the American Board of Surgery (for the individual), and the RRC (for the program), are driven by numbers. I think that Dr Aust was correct, even though my agreeing with him will destroy the basis of our relationship. What the American Board of Surgery is doing and what the RRC is doing under the leadership of the Accreditation Committee for Graduate Medical Education is looking at other aspects of what exemplifies a surgeon. One of those is a new measurement called competence. This is going to be a hot item over the next year or two, first trying to define it, as it has been defined in industry for its workers. Competence eventually will be defined for a surgeon's performance. Competence has to do with how we apply our experiences, what our judgment is, what our integrity is, what our basic culture is. I am encouraged as I see operations change because I think that is good for patients; and I am also encouraged to see that we are teaching surgeons to care for patients if they operate on trauma or not. The care of the critically ill was beautifully described in our presidential address yesterday. All of the things that surgeons do are not just procedures. I really appreciate a paper like this being on the program.

Jeffrey H. Peters, MD, Los Angeles, Calif: It's easy to assume, I think, that these operations increased or decreased because of the changing volumes in surgery as it is practiced in the United States over the years. This may not be a valid assumption. I wonder if Dr Parsa would focus on the reasons for the changes, specifically, are there other biases in play here? That is, are the residents keeping better data, or probably more importantly, are there increasing trends in the transfer of the operation from the faculty physician to the resident, which may be a role in the increase in operations like Whipple operations and in the decreasing prevalence of operations like antireflux surgery.

Lawrence W. Way, MD, San Francisco, Calif: To what extent are the changes you are reporting due to local referral patterns or to genuine trends in surgical practice? For example, have you compared these data from your residency program with the national data from the RRC for surgery?

There is growing evidence that it takes more than just well-developed general skills to do the more complex operations in general surgery. The literature on this subject indicates that the specific procedures must be rehearsed regularly for competence to develop and be maintained. This poses a significant challenge to the current system of surgical education considering the profession's fiduciary obligation to the public to provide competence along with certification.

Theodore X. O'Connell, MD, Los Angeles: I have several questions and comments. The first is that on both lists, the early list and the later list, peritoneovenous shunts are in the top 10. That is a little hard to believe, and I think that it may be miscoded and that these are really vascular access procedures. These are Hickman, Broviacs, or renal dialysis access procedures rather than peritoneovenous shunts. I wish the authors would comment on that.

My next 2 comments are from the perspective of a program director for more than 20 years. I think we have become very number driven, even more number driven in the last couple of years with the operative logs being computerized. I think that being number driven has its good points, but also its bad points. I think some of the increase in numbers is simply because we are number driven. As a program director, whenever one of my residents is low in number on a particular procedure, I make sure he or she gets it one way or the other. The RRC ups the numbers and then we up the numbers. So we keep on chasing after that as they keep on reviewing it. So it's almost an endless procedure of chasing our tail. I don't think it's necessarily a good thing.

Second is a message to the RRC. They really have to look at some of these procedures and change them, because if you are a cutting-edge program, and I use "cutting-edge" in the best way, sometimes you will be "dinged" by the RRC because you are not doing enough of the cases that they expect because you are more modern and doing it by noninvasive methods or whatever. I remember about 10 years ago, we got dinged for not having enough subphrenic abscesses, which I thought was a hard thing to realize, but we actually were warned about this. I didn't know if I should be proud or ashamed for not having too many of them. The other thing I think reflects what Dr Aust said is that we have no way of measuring surgery not done. Judgments, deciding which patient is really not a candidate for an operation, discussion with your attending and so forth, in deciding when not to operate should be mentioned too, and those numbers should be reviewed too as well as the actual operations done.

Tom R. DeMeester, MD, Los Angeles: The comment I am about to make is a difficult one for me to make, and I will probably wish I hadn't made it. I think we have to begin to seriously look at the trends that are occurring. The list of most common procedures performed by our residents consists of very few sophisticated procedures. That's a problem. When our graduates have performed on average 1 gastrectomy, 1 esophagectomy, and 2 Whipple operations during their residency, are they ready to perform these procedures on the public? I don't think so. As the body of surgical knowledge expands, we just can't cover it all. I wonder if the time has come when we must seriously take stock of what we are achieving. What we are accomplishing is training in the basics of surgery. Our residents are graduating with skills in just basic surgery. They are not ready for sophisticated, complex operations. Maybe we should adjust our teaching program to something like a 3-year program in basic surgery, followed by 2 years focusing in specific areas such as colorectal, vascular, critical care, or hepatobiliary or whatever. Maybe this would be a better way to go about it in the new millennium. Please don't crucify me for making such a suggestion, but we do not want to compromise the future of surgery by persisting in doing what we cannot or should not do.

James E. Goodnight, Jr, MD, PhD, Sacramento, Calif: The RRC must have known that this paper was to be presented today and therefore put Dr Organ on recall to Oakland. The RRC is in fact doing a regular multiyear review of the East Bay Program today and we felt that Dr Organ should be there to meet and great the reviewers as only he can.

I will attempt to close and must emphasize the work presented is that of Drs Parsa, Organ, and Barkan. I certainly appreciate the comments of Dr Aust, who has a remarkable perspective on surgical education. I couldn't agree more and other discussants brought up the need to look at the nonoperative work that the residents do; the judgment that comes from nonoperative trauma management and critical care management is a key piece of surgical maturation.

Dr Gewertz asked if fellowships actually bring more and better cases that become available for general surgical residents, and what the effect of fellowship is. This wasn't evaluated by the study, but it is the perception of the authors and Dr Parsa that in fact the presence of sophisticated fellowships does enhance the number of cases and does enhance the general surgical experience.

Dr Prinz asked if the number of cases has increased. Yes, the number has increased, not only in the period of the study, 1991-1997 (the 6.8% increase that Dr Parsa mentioned); actually, the number of cases has increased steadily since 1984. So the authors would say at least at this point that there is a more intense operative experience in training. If I understood Dr Prinz's second question correctly, in the period of the study, there was an increase in parathyroidectomy and thyroidectomy, suggesting that there is an influence of subspecialists. Did this change measure the subspecialization of faculties? Again, the speculation would be yes, but it cannot be directly addressed by this study.

Dr Shuck mentioned the issue of competence, an issue that I am sure is debated every time the American Board of Surgery meets and every time that the American Board of Surgery sits down to evaluate an examination. Clearly, it will be a major issue for this profession over the next 2 or 3 decades in terms of how we present competence to the public. Once again, we would agree that the nonoperative experience must be carefully evaluated.

Dr Peters asked if there are other factors and if the residents are reporting better. Yes, absolutely. Anyone who has been involved in resident training knows that the overall reporting mechanisms have steadily evolved over the last decade and undoubtedly that has had an influence. Do the figures correlate with attendings handing off more cases? That couldn't be answered from this study.

Dr Way asked if Dr Organ compared these data with his local program. The answer is yes. I happened to see these numbers and it is interesting. The national experience does correlate within the microcosm of their residency, which is quite a sizable residency, as well as with our residency in Sacramento.

Dr O'Connell had several questions. One was whether peritoneovenous shunts were too high in the data. Absolutely, and we were totally surprised by this. It doesn't reflect our experience. It probably has to do with this reporting category on resolution. Do the numbers drive the requirements of the RRC? Almost certainly, but we cannot speak to that. Then he mentioned the problems with the RRC review of his program. All I think we can say is that life is hard, Ted.

Dr DeMeester raised the critical philosophical question. Are the residents graduating with these numbers competent? Perhaps this is another way to ask if the health maintenance organizations want people who can do a few operations extremely well, very efficiently, and most cost-effectively? All of the papers reporting these days indicate that the more times you do an operation, the better you do it. That we all know quite well. What do we do about that in training? It is an interesting and I don't think an opposing view, but a parallel view that there is the need to train surgeons very broadly, both in operative technique and nonoperative judgment. That foundation, that maturation has to occur. When do you go for the gold, the absolute efficiency, the absolute numbers? It is hard to say. There was a very interesting paper presented at the American Surgical Association by Ritchie last spring about surgeons applying for recertification and the breadth of their experience. They demonstrated a remarkable breadth of experience ranging across the whole field of general surgery. The perception that we get, I think, not just from that paper, but as surgeons involved in training programs, is that there is a remarkable competence out there. There needs to be a cadre of people who are extraordinarily well trained in highly technical procedures. I suspect we will deal with it just like we have with the other problems in general surgery, and I suspect that we will deal with it very well.