Solomon H, Chao AB, Weaver FA, Katz SG. Change in Practice Patterns of an Academic Division of Vascular Surgery. Arch Surg. 2007;142(8):733-737. doi:10.1001/archsurg.142.8.733
Technological advances have required that faculty of academic divisions of vascular surgery acquire new technical skills and significantly alter their past clinical practice patterns.
Retrospective medical record review.
An academic tertiary referral center and a community teaching hospital.
All patients undergoing 10 specific vascular procedures during a 5-year period.
Main Outcome Measures
We analyzed volumes for 10 specific open and endovascular index procedures performed by 5 vascular surgeons during a 60-month period. Procedures reviewed included open abdominal aortic aneurysm repair, endovascular abdominal aortic aneurysm repair, carotid endarterectomy, carotid artery stent, suprainguinal arterial reconstruction, suprainguinal percutaneous transluminal angioplasty/stent (PTA/S), infrainguinal arterial reconstruction, infrainguinal PTA/S, renal and visceral arterial reconstruction, and renal and visceral PTA/S. In-hospital length of stay was compared between open procedures and their endovascular counterparts.
In 2000, 453 open and 44 endovascular index procedures were performed. In contrast, by 2005, open index cases had decreased by 47.0% (239) and endovascular index cases had increased by 679.5% (299). Open abdominal aortic aneurysm repairs had decreased by 54.5% (68 vs 31), carotid endarterectomies by 28.8% (139 vs 99), suprainguinal arterial reconstructions by 47.5% (40 vs 21), infrainguinal arterial reconstructions by 56.5% (186 vs 81), and renal/visceral arterial reconstructions by 65.0% (20 vs 7). In 2005, 62 endovascular abdominal aortic aneurysm repairs and 45 carotid stents were performed, whereas none were performed in 2000. In addition, infrainguinal PTA/S had increased by 675.0% (12 vs 81) and suprainguinal PTA/S by 20.0% (20 vs 24).
Although the total number of procedures performed has remained relatively constant, there has been a dramatic increase in the number of endovascular procedures as well as an associated decline in the number of open procedures. This change in practice pattern has allowed the members of our division to maintain a significant role in the care of patients undergoing vascular surgery, as evidenced by stable overall procedural volume. This will provide a platform for future outcome-related analyses of open vs endovascular procedures performed within a single specialty group.
The use of the aortic homograft by Gross et al,1 Oudot,2,3 and Dubost et al4 in the mid-20th century began the modern era of vascular surgery. Development of fabric grafts by Voorhees et al5 and DeBakey et al6 allowed abdominal aortic aneurysm (AAA) repair to become routine. The pioneering work of Dos Santos7 and Leriche8 in endarterectomy formed the basis for successful treatment of arterial occlusive disease. Adaptation of Carrel's suturing technique to arterial bypass by Kunlin9 led to the treatment of infrainguinal occlusive disease using vein bypass. The landmark reconstructions of the carotid bifurcation by Eastcott et al10 and DeBakey11 in the 1950s made the treatment of extracranial carotid disease commonplace. These procedures established open surgical techniques as the gold standard for the treatment of peripheral vascular disease.
More than 50 years ago, Seldinger12 developed the technique of femoral artery catheterization used today. In 1964, Dotter and Judkins13 performed the first successful arterial dilatation. Ten years later, Gruntzig14 created the modern angioplasty catheter. In 1990, Palmaz et al15 published their initial experience with the deployment of balloon expandable stents, ushering in the modern era of endovascular management of arterial disease. Researchers such as Fogarty et al,16 Greenfield et al,17 and Parodi et al18 have played significant roles in the minimally invasive treatment of arterial embolus, pulmonary embolism, and aortic aneurysm. However, those in disciplines other than vascular surgery have advanced most developments in the endoluminal treatment of arterial disease. With the ever-increasing demand for minimally invasive procedures, members of our vascular surgery division retrained to develop catheter-based skills. In 2000, we began to perform minimally invasive vascular procedures.19 This article describes the evolution of our practice during the ensuing 5 years.
We reviewed the records of all patients who had undergone any of 10 open or endovascular index procedures performed by the 5 members of our division during a 60-month period. Patient data were collected from inpatient and outpatient medical and billing records. Patient demographics and comorbidities were recorded, and comparisons were made between those undergoing minimally invasive procedures and those undergoing open repair. Procedures reviewed included open AAA repair (ruptured AAAs were excluded), endovascular AAA repair (EVAR), carotid endarterectomy (CEA), carotid artery stent (CAS), suprainguinal arterial reconstruction, suprainguinal percutaneous transluminal angioplasty/stent (PTA/S), infrainguinal arterial reconstruction, infrainguinal PTA/S, renal and visceral artery reconstruction, and renal and visceral PTA/S.
The volumes of each of these procedures, as well as their corresponding in-hospital lengths of stay and mortality rates, were recorded. Length of stay was defined as postoperative days until discharge or completion of care by the vascular service. Postoperative mortality was considered to occur within 30 days of the procedure or during the patient's procedural hospitalization. Comparisons of case volume, length of stay, and perioperative mortality rates were made between the first and last years of the study. Vascular procedures performed by the general surgery residents in our institution in 2000 and 2005 were recorded, as were the number of open and endovascular procedures performed by the vascular surgery residents. Board examination passage rates for general and vascular surgery residents during the study period were noted. The Mann-Whitney test was used to determine differences between variables with continuous outcomes, and the Fisher exact test was used to analyze differences between the categorical variables. Statistical significance was assumed at P < .05. Data analysis was performed using SAS statistical software (SAS Inc, Cary, NC).
In 2000 and 2005, the University of Southern California Division of Vascular Surgery and Endovascular Therapy performed 1035 index procedures. Of the patients, 62.9% were male and 37.1% were female. Their ages ranged from 49 to 92 years (mean age, 76.8 years). In addition, 45.9% of the patients confirmed a history of smoking more than 20 cigarettes per day, 40.9% had hypercholesterolemia, 37.0% had coronary artery disease, 20.9% had diabetes mellitus, 7.8% had chronic obstructive pulmonary disease, and 5.0% had renal impairment. There were no significant differences in the demographics or comorbidities of patients operated on in 2000 and 2005 or in patients undergoing open or minimally invasive repair (P>.16).
In 2000, 453 open repairs and 44 minimally invasive procedures were performed for the 10 index cases. In contrast, by 2005, 239 open repairs and 299 minimally invasive procedures were completed. By the final year of the study, CEAs had decreased by 28.8%, open AAA repair by 54.5%, suprainguinal arterial reconstructions by 47.5%, infrainguinal arterial reconstructions by 56.5%, and renal and visceral arterial reconstructions by 50.0%. There was a corresponding increase in the number of minimally invasive procedures performed (Table 1).
When compared, patients undergoing minimally invasive procedures had significantly shorter hospital stays than those having open repair (1 day vs 3 days; P<.001). This held true when all endovascular procedures were compared with their open counterparts (Table 2). Patients undergoing CEA had a much greater variation in the range of their length of stay (range: 1-19 days vs 1-8 days). This caused a statistically significant P value even though the median values for CEA and CAS were equal. Mortality data were available for 97.2% of patients (1006/1035). The 30-day mortality rate was 0.6% for minimally invasive procedures and 2.2% for open surgical repair (P = .09) (Table 3).
In 2000, general surgery residents completing their training in our institutions performed an average of 83 defined-category vascular cases, and in 2005 they performed 64. In 2000, the vascular resident performed 179 open and 38 endovascular procedures, and in 2005 he performed 203 open and 110 catheter-based procedures (only the 10 index cases were tabulated). During the study, 90% of the general surgery residents passed the American Board of Surgery qualifying examination on their first attempt, and 86% successfully completed the certifying examination on their first effort. First-attempt board passage rates for vascular surgery residents were 83% for the qualifying examination and 100% for the certifying examination.
Surgery has rapidly evolved in the last 2 decades. Laparoscopic techniques have revolutionized the specialty, and robotic and endoscopic surgery hold promise. Surgeons have been required to learn and utilize new skill sets to keep pace with the advancing technology. Demand for endovascular procedures is rapidly increasing as well. This will oblige vascular surgeons to acquire catheter and guide wire skills or entrust the care of many of their patients to physicians in other specialties. Members of our division chose to retrain and acquire the requisite skills necessary to provide our patients with catheter-based treatment options. As a consequence, our practice patterns have been dramatically altered.
The results of our study demonstrate the rapid increase in minimally invasive procedures that has occurred during the last 5 years. The number of EVARs and suprainguinal and infrainguinal PTA/Ss has now eclipsed the number of their open surgical counterparts. The CAS is beginning to replace traditional CEA as the treatment for patients with extracranial cerebrovascular disease. One result of this change in practice patterns is a dramatically reduced length of stay for those undergoing minimally invasive procedures. Many of these procedures can be performed on an outpatient basis. This makes inpatient beds, which are increasingly becoming a valuable commodity, available, and lowers the cost of patient care.20 When compared, a reduced mortality rate for endovascular procedures approached but did not achieve statistical significance (P = .09).
New technologies need to be critically evaluated, and their efficacy needs to be objectively compared with traditional treatment modalities. We believe that surgeons who can perform both minimally invasive and open procedures will be better able to make these unbiased judgments than physicians who provide only 1 form of care. In addition, those who are facile in all modes of therapy will be able to offer their patients the best therapeutic option after considering all possible alternatives.
Postgraduate training will also be greatly influenced by rapid technological advances. Some of these new techniques will require the acquisition of skill sets that traditionally lie within other specialties. It is incumbent on those who train residents to learn to skillfully perform these new procedures so that they will be able to help their students appropriately incorporate them into their therapeutic armamentarium and treatment algorithms. Realizing the importance, utility, and prevalence of minimally invasive procedures, our residency review committee has now required all vascular surgery fellowships to incorporate endovascular procedures as an integral part of the curriculum. Similar requirements have been mandated for general surgery residencies in the areas of laparoscopy and endoscopy to provide more comprehensive and up-to-date training for the next generation of general surgeons. Surgeons who integrate these new techniques into their practice will not be required to delegate the training of their house staff to those in other disciplines, who may or may not inflict their own treatment bias. They will also serve as role models for future surgeons, who will most likely be required to retrain several times during their careers. Another consequence of the rapid rise in endovascular procedures is the resultant decrease in traditional open vascular procedures available in general surgery resident training.21 Currently our programs graduate 7 to 8 general surgery residents and 1 vascular resident per year. Although the number of vascular procedures performed by the general surgery residents has decreased, to date all have exceeded the minimum requirement set forth by the American Board of Surgery. However, the American Board of Surgery may need to reevaluate resident case requirements in the future. Board passage rates for both general and vascular surgery residents have remained satisfactory. Since increasing the length of our vascular surgery residency in 2002 from 1 to 2 years, all vascular surgery residents completing our program have been able to receive privileges for all open and catheter-based procedures with the exception of CAS. We plan to rectify this deficiency in the near future.
This study demonstrates a dramatic change in the practice pattern of 1 group of vascular surgeons. We believe that surgeons are best able to care for patients with peripheral vascular disease. The acquisition of catheter and guide wire skills has allowed the members of our division to maintain a steady practice volume without erosion of our patient base. If we had not incorporated these new techniques into our practice, we would likely have relinquished the care of a large portion of our patients to physicians in other specialties.
We conclude that rapidly changing technological advances will greatly influence the practice patterns of physicians, now and in the future. The acquisition of new skill sets, often outside the bounds of traditional surgical practice, will be required if surgeons are to keep pace. A proactive approach to learning will allow surgeons to offer the best available treatment options while maintaining their patient base.
Correspondence: Steven G. Katz, MD, Office of Medical Education, Huntington Hospital, 100 W California Blvd, Pasadena, CA 91105 (firstname.lastname@example.org).
Accepted for Publication: March 16, 2007.
Author Contributions:Study concept and design: Chao, Weaver, and Katz. Acquisition of data: Solomon, Chao, and Weaver. Analysis and interpretation of data: Solomon, Chao, Weaver, and Katz. Drafting of the manuscript: Solomon, Chao, Weaver, and Katz. Critical revision of the manuscript for important intellectual content: Solomon, Chao, and Katz. Statistical analysis: Chao. Administrative, technical, and material support: Solomon and Chao. Study supervision: Solomon, Weaver, and Katz.
Financial Disclosure: None reported.
Previous Presentation: This paper was presented at the 78th Annual Scientific Meeting of the Pacific Coast Surgical Association; February 20, 2007; Kohala Coast, Hawaii; and is published after peer review and revision. The discussions that follow this article are based on the originally submitted manuscript and not the revised manuscript.
Additional Contributions: Gabriel Akopian, MD, Haimesh Shah, MD, and Erin Atkinson, BA, provided valuable help with acquisition and analysis of our database.
Cornelius Olcott, MD, Stanford, California: Many years ago when I was preparing to give my first paper, which incidentally was before this association, Dr William Blaisdell told me that you can judge the success of a paper by how much discussion it engenders. I anticipate that this paper will stimulate significant thought and discussion.
Dr Solomon and his colleagues describe the changing practice patterns in an academic vascular surgical practice over the past 5 years. It should come as no surprise to any of us that the practice of vascular surgery, like all surgical specialties, has changed significantly since 2000. Less invasive procedures have been created for the repair of AAAs, carotid stenosis, aorto-iliac and lower extremity occlusive disease, and renal and visceral occlusive disease. These are the index cases for this study. This trend toward increasing numbers of endovascular cases, documented in this paper, is being noted across the country; certainly we see this at Stanford. These endovascular procedures are an important addition to the armamentarium of the vascular surgeon, and it is important that established surgeons and trainees become well versed in their use.
While this paper does not impart any new or unexpected findings, I believe that it serves an important purpose in that it will stimulate discussion among surgeons and educators as to how best to deal with this change in clinical practice. To this end, I would like to raise several questions and issues for the authors to address.
First is the issue of training. The authors correctly point out the need for surgeons to be open to retraining to keep up with advancing technologies. There is no question that this is true and important. However, the authors do not spell out what constitutes adequate retraining. Does this require a full fellowship, a “mini-fellowship” (typically 3 months), or simply on-the-job training by a partner or colleague who has been trained in endovascular procedures? This is an issue presently being debated in the Credentials Committee of my hospital. It has become apparent that what vascular surgeons consider adequate retraining may not be acceptable to interventional radiologists and cardiologists. What do the authors consider to be the optimum training for the practicing vascular surgeon? Also, how should our present vascular trainees be trained in catheter skills: only by other vascular surgeons, or should physicians from other disciplines be included? For example, most vascular surgeons are pretty good at the endovascular management of iliac occlusive disease and renal artery stenosis. However, only a few have significant experience with complex problems such as fenestration procedures in acute aortic dissections or complex arteriovenous malformations. How should our trainees gain experience with these more difficult disease entities? In addition, how will all these changes impact the training of general surgery residents? This is a concern, I am sure, to many members of this association.
A second issue is quality. While I agree that often the less invasive procedure is preferable and equally efficacious, we are now learning that this is not always the case. The authors state that the CAS is beginning to replace endarterectomy. Indeed, many vascular surgeons now preferentially do angioplasty and stenting. However, as the data come in, we are learning that CAS may not be as good as TEA (thromboendarterectomy) in many patients (eg, those older than 80 years to whom one would like to offer a less invasive procedure). Also, 2 European trials were recently discontinued because the results of CAS were not as good as anticipated. Certainly, the endovascular treatment of AAAs (EVAR) is accepted treatment for some AAAs. However, there are some recognized downsides to EVAR: patients have to be followed closely, frequently with CT (computed tomographic) angiograms, which are expensive, are frequently inconvenient, and expose the patient to significant radiation. Also, there is a significant reintervention rate with EVAR, which is not seen with open repair of AAAs. This paper, like many papers on endovascular techniques, does not take into account the cost, inconvenience, and potential morbidity of the required follow-up and reinterventions. We need better outcome data for both open and endovascular procedures so that we can provide our patients with the best treatment for their particular situation.
A third concern is what is driving this trend. I believe most surgeons want to do what is really best for the patient. However, there is no question that the quest for market share and the desire to increase compensation enter the picture. In the vascular arena, the endovascular cases typically pay better and do not carry the 90-day global coverage common with open cases. The medical industry also is pushing this trend. We have all been exposed to advertisements on TV and in the newspapers appealing directly to patients to “ask their doctors” if a particular drug or device is right for them. This encourages patients to push for the new technology. It is hard for even the most honest of surgeons not to consider these factors when choosing a procedure. How do surgeons at USC ensure that any given patient receives the optimum treatment for his or her particular situation, and how do they deal with the differences between open and endovascular reimbursement in their compensation program?
Lastly, I think it will be interesting to see where this all ends up. I doubt that vascular surgery will be the same 5 to 10 years from now. I would not be surprised to see the vascular surgeon of the future be either a traditional open surgeon or an endovascular specialist. I believe it will be increasingly difficult to maintain optimum skills in both areas. Unfortunately, no one knows what the future will hold. However, Dr Solomon and his group have certainly given us food for thought.
Dr Katz: I will try to answer your questions in order. With regard to the first question about training, I think there are several ways that this can be accomplished. Taking a mini-fellowship for 3 months is one option. However, I think it is very difficult for practicing physicians to take 3 months out of their practice to participate in one of these fellowships. The hospitals in which we practice formed interdisciplinary committees who chose to adopt the American Heart Association standards of 100 diagnostic and 50 therapeutic interventions for endovascular privileging. We were fortunate to have one of the members of our group fully trained in endovascular techniques. He was kind enough to mentor us through all of those proctored cases. We have previously validated in a recent manuscript that these privileging criteria, if adhered to, can help to ensure quality.
How best to train the residents? We feel that it is most efficacious for vascular surgeons to train general and vascular surgery residents. I'm not sure that physicians in other disciplines feel quite the obligation to train our residents that we do. I also do not think that we as vascular surgeons are required to perform every endovascular procedure. We often ask our colleagues in interventional radiology to perform thrombolysis and difficult embolization procedures. We have learned from them, and they in turn have learned from us. Although times were difficult at first, we now have a very collegial relationship with our interventional radiologists.
The increase in endovascular surgery is sure to eventually impact general surgery training. The number of open vascular cases is clearly decreasing. Unfortunately, these are the cases that are most appropriate for general surgery resident training. Presently we are fortunate enough to be able to provide sufficient case numbers for both our general surgical and vascular residents. I am not sure, if the trends continue, that we will be able to do so in the future.
I think first we learn what we can do and then later what we should do. Carotid artery disease provides an excellent example. We are learning that the octogenarian with symptomatic carotid stenosis is probably best treated with surgery rather than stenting. I think as time goes on, we who can provide both open and endovascular therapies will be best able to make accurate decisions as to what form of treatment is best for patients. There is an old saying that when your only tool is a hammer, everything looks like a nail. For many of the other specialists who are able to use only 1 treatment modality, that tends to be true. I believe that surgeons who are able to offer all therapeutic options will be much more objective, not only in treating their patients but also in evaluating new technologies.
As mentioned, some of these newer techniques do require more frequent follow-up than traditional procedures. We can expect that technologies will continue to improve. As time goes on and the devices become more sophisticated, there will most likely be less of a need for frequent evaluation and subsequent interventions.
As Dr Olcott pointed out, we are now facing pressure from both patients and the industry to perform certain procedures and to employ new technologies. It will be a continual struggle to resist these forces, but we must remain firm in our commitment to providing the best care for our patients. We have not considered differential reimbursement to be an important part of our treatment algorithms. While the maintenance of market share is attractive, the ability to deliver optimal treatment to patients with vascular disease should remain paramount. We believe that vascular surgeons trained in both open and percutaneous techniques can best deliver this type of care.