| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1999;68:374-376
© 1999 The Society of Thoracic Surgeons
Presented at the Thirty-fourth Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 26–28, 1998.
DR ALFONSO CHISCANO (San Antonio, TX): Thank you so much, Fred, for letting us get into the new era. I have a few comments. If you train your residents well and you kick out those who are not adequate, you are going to have a nice production of people who can do these operations and this care in an appropriate way.
Managed care and peers are present in this issue. The friendly use of these data is what we need to get. I’m so surprised to see these data, which is not quite appropriate in the hands of people who try to punish you either economically, or your peers say, "Well, maybe I should do this case instead of you." For the decisions of catheterizations and going to surgery, you don’t get into that data. There is no way that you can get these cardiologists in line with you to be able to place these things into the proper perspective. When you get a patient with a cardiologist pumping on the chest and they know in their heart that the patient is not stable, you don’t get these people inside of any of those risk factors.
Dr Dziuban cut his mortality from 4.5% to 1.5%. What did you really do, Dr Dziuban, to do that? If you don’t have the autopsy of your patients, you don’t really know what the patient died from. I remember 10 years ago I was encouraged to get an autopsy but I then got a friendly letter from a lawyer, "Hey, the patient died of something wrong with the spleen. I’m going to sue you." So probably that hid the truth of the real cause of death.
Dr Sergeant, you have a group of your patients with a 48% mortality. If I operated with that mortality, my friendly peers, the sharks, would throw me in the tank. I have been in Europe many times. I organized a program in the Canary Islands in 1982 where we operated, I think it was four open-heart surgeries for four coronaries. I went there last week. They are doing 600 cases a year. They are getting paid, for everything, hospital and everything, $8,000. Your follow-ups are fantastic. I never will get my cardiologists to take care of our follow-ups ever. Perhaps what we should do is just take lots of our Medicare patients to the Canary Islands, fix it, and make ourselves $10,000.
Thank you so much.
DR DZIUBAN: First, I have to say that these are not my cases and my mortality. I can’t take credit for that work. My associates can. But you asked how we reduced the mortality specifically from 4.5% to 1.5%. It wasn’t by any specific single change. It was a spectrum of changes throughout the whole program. Within surgery, the primary changes were probably changes in cardioplegic technique. With better coordination with cardiology, there was a change in our use of the intraaortic balloon. There were changes in the nursing department and development protocols for nitroglycerin. We found in these high-risk patients that we were using the intraaortic balloon infrequently. The cardiologists were pushing to get these patients in to be revascularized immediately but they needed a little more stabilization. A little delay for stabilization and many other factors produced these improved results.
You mentioned that you can’t get the cardiologists doing things the way you would. As hard as it is to get the data and work with it and interpret it, that’s still the easier part of it. The most difficult part is bringing together specialists who are used to looking within their own specialty primarily and doing what seems to suit their needs as opposed to looking at perhaps the overall needs of their patients. A cardiologist who looks at the surgeon’s needs and thinks about how to prepare his or her patient for surgery is a much more valuable asset to you as a surgeon, and optimizes the care of the cardiac surgical patient. That kind of integration is essential, I think, to reduce mortality beyond a certain point because each specialty can only do so much by itself.
DR SERGEANT: I have two comments. First of all, you pointed out in the patient spectrum of the 30-day mortality some patients with very high mortality. If you treat patients in cardiac massage, who are at the high spectrum of age, with bad ventricular function, for whatever reasons they are in cardiac massage, if you start calculating these risks in these patients, that’s precisely what it’s all about. It is not about the fact that coronary surgery has a 2% risk. Whoever says that coronary surgery has a 2% risk has no insight into the data.
The point is that you can do thousands of patients with 0.5% and less risk, but a certain small subgroup of patients with an accumulation of risk factors will get you very high risk, and that’s what the accountability is all about, correcting for this variability in patients.
Now, the second point, you mentioned that the cardiologists should take care of the follow-up. I can tell you that, in my opinion, it’s the responsibility of whomever is treating and taking care of the patient. The way we have structured this is around the cardiologists because that’s their primary target, but it’s the surgeon’s responsibility to know what is happening with the treatment and with the patients.
DR GROVER: I think that brings up a good point also. What we measure is one interval in the long-term treatment of someone with, in this case, ischemic heart disease. This is one reason for the efforts of Bruce Ferguson and Laurie Shroyer on behalf of our STS Database Committee to collaborate with the American College of Cardiology to eventually get them on board to collect this type of joint data. It will be very enlightening for them, too, as they examine the data for high-risk patients when considering referring them for surgery, and may cause them to consider some alternate treatment.
DR JOHN R. BENFIELD (Sacramento, CA): This was an enormously important and illuminating presentation. We heard, among other things, Bill Nugent’s report of looking at six different institutions, and if I heard the numbers correctly, 8,000 procedures during a 1-year period of time. I would like the panel to address itself to the difference in practice patterns between cardiac surgery and general thoracic surgery. We know that practice patterns in general thoracic surgery are very different from cardiac surgery. I venture a guess that if Bill had looked at those same six institutions in New England with regard to general thoracic procedures, that there would not have been 8,000 cases of general thoracic surgery. I would guess that in New England there probably are many hospitals in which 20 and 30 general thoracic surgical procedures or perhaps even fewer are done in 1 year. So the question is how the general approach that you have outlined so beautifully today can practically be adapted to general thoracic surgery?
DR NUGENT: One size need not fit all. As we work to create and implement a general thoracic surgical database, we need to keep sight of what outcomes need to be followed in this population. Whereas mortality has become the de facto agreed-upon metric in cardiac surgery, more subjective outcomes might be followed in the general thoracic surgical population, for example, functional health, dyspnea, or long-term survival.
I reiterate the message that these clinical databases must be seen as tools for better understanding and improving our patient outcomes. In the thoracic surgical population, we must ask ourselves which outcomes are most important. We may be surprised to find that it isn’t the mortality rate alone.
DR DZIUBAN: I think this is an excellent question. The bottom line, I think, is that general thoracic surgery is such a different animal that, while we can take some of the lessons from our cardiac surgery database, the end product is bound to be very different. I firmly believe that in today’s world, as a surgeon, what we need is a database that meets all of our needs rolled into one. We can’t afford to keep separate tallies of things. And in thoracic surgery, just to enumerate a few of the problems, unlike cardiac surgery, first of all, the work is very diverse. The cardiac surgeon can have a cardiac database that encompasses most of his work. Eighty percent of it might be one category or procedure, and the risk models and tallies fit very well with their overall volume. In general thoracic surgery, the diversity of work makes it difficult because if you build a database to cover the diversity of work to look at the overall picture, then it makes it difficult to get into the details of each. So you need subportions databased separately, as we are developing a lung cancer database, esophageal cancer, etc. In addition, there are different goals often in thoracic surgery. For example, some procedures often are more palliative or diagnostic than in cardiac surgery. And finally, there is a wide diversity of practitioners, many of them outside of our own specialty. I still think that the best we can do is to try to move forward with this very complex situation and build the best database modeling we can, to know where we stand in our specialty, and then potentially begin to compare with others outside of it.
DR JOHN R. HANDY, JR (Portland, OR): I would like to ask Dr. Nugent, with a regional database that enters 8,000 patients a year, who pays for the entry, the maintenance, and the evaluation of that data? How do you fund it?
DR NUGENT: Believe it or not, for the first 6 years of our experience, the Northern New England Consortium was completely funded internally by patching together small research grants. We now estimate that the actual database costs about $300,000 a year to maintain. This calculates to about $30 for each chart that is entered. That is less than 1% of the total cost of the coronary bypass operation. Currently, we are funded in part by dues coming from each institution, as all data that are submitted are protected under quality assurance laws. Most institutions have incorporated this cost into their quality assurance budgets. I believe that each organization has worked out a different strategy for getting data into the database. Each data form must be complete and is subsequently validated for both the outcome and procedure done. In our Section of Cardiothoracic Surgery at the Dartmouth-Hitchcock Medical Center, I have devoted a portion of one full time equivalent employee to maintaining our sectional database. All of our data are initially collected on paper data forms and subsequently key punched into a computer. We found that key punching was problematic and, therefore, had to outsource the key punching in order to keep our database up to date. This costs my section approximately 50 cents a chart, or about $300 a year. We do not feel that these expenses are excessive.
DR DZIUBAN: I have a quick comment that I think might interest you about a phenomenon that we have observed in our practice, which covers three hospitals. We initially started out several years ago doing all the database information collecting and processing, and submitting it ourselves through our practice at our expense, primarily to keep control of it and assure the quality of the data, which reflects our care. What has happened over the years, as the hospitals have become more aware of the need and value of this type of data from their own standpoint, is that they have taken a very active interest in this, and we have been able to shift this effort to them, with appropriate controls to assure that it is being done in the way that we set it up.
When partnering with a hospital, they often are willing to share in the expenses since they need the data as well.
DR ROBERT GIANGUZZI (Thornton, PA): Our institution recently hired a physician trained in medical informatics to help us along with our cardiac surgery and invasive cardiology database. While he points out that we’re heading in the right direction with the clinical endpoints, data integrity, and all other database development principles, he wants to know if we should take a look at the prehospital phase and all the business transactions, the clumsy business transactions that occur from precertification to delay in primary care referrals, resulting in delays to access for cardiac tertiary procedure. Should we be looking at that information as well?
DR GROVER: That’s a very good question. I think it involves a question of, should we be measuring access to care as well as the nuisance factor of all of these verifications of insurance? We’ll keep that in mind. As I mentioned, access to care is something that does concern us. We are caught between trying not to let this database get too big in terms of the number of variables, but also expanding into such things as access to care and long-term follow-up.
DR NILES D. CHAPMAN (Los Angeles, CA): One of the curious things in Los Angeles County in an area called the San Fernando Valley is that there are eight hospitals that do cardiac surgery and only one of those is in the database. The question comes up, are many of the hospitals self-selected that are in the database? The question comes up, how representative is the data of the overall practice of cardiac surgery in the United States?
DR GROVER: Whenever you have a voluntary database, there is always the possibility that you are selecting out a group of institutions that are willing to share their data. And are those institutions’ results different from the group as a whole? We think, but we cannot prove, that we have a relatively good cross section since we have something like 70% penetration of the cardiothoracic surgeons. But that’s always a question. I would guess that our penetration is high enough to where selection is probably not an issue.
DR THEODORE G. PHILLIPS (Allentown, PA): Given the 10-fold increase in mortality with salvage operations for coronary artery bypass grafting, and the probable 10-fold increase in cost to take care of that patient and also the probable 10-fold decrease in the quality of life afterwards, hospitals are starting to clamp down and giving innuendos not to operate on these people because it incurs such a tremendous cost. My question is, are we going to get guidelines from The Society on when to say no? Nowhere in the databases are you penalized for saying no. As you know, you take a hit to say no in many respects with referrals and with reputation, but "no" doesn’t appear in the database. "Yes" does appear in the database, and it’s a big cost. Are we really providing a service to these patients to do them? It’s hard to say no to someone if you have a chance to possibly save them, but, on the other hand, if we don’t have the follow-up to see what quality of life these people have, what are we doing to these people? And if we don’t get direction from the Society, the health care providers and the hospitals are going to dictate to us who we should operate on and who we shouldn’t, and we’re going to ration care.
DR GROVER: The question is whether we should get involved with practice guidelines, and I’ll make one stab at it and then let Paul answer. I would say that we probably should let the data speak for itself, and I would think if we show the cost and the mortality related to this, then it’s probably best that we leave the discussion of whether or not to operate to each surgeon and his or her team’s independent judgment.
DR SERGEANT: We have followed hundreds of patients in severe shock and in cardiac massage at the time of surgery, and we can very well address your issue. Precisely, if you look into the follow-up, you identify that the mortality is extremely high, as has been pointed out by one of the discussants, but you can identify that the hazard really over time then reduces dramatically and the 10-year survival is around 50%. So this means that you have a high early price, but once these patients go through this early phase, the quality of life basically is very good, and that’s precisely what outcome analysis is all about, that we shouldn’t point that much to the early events or the early costs; it’s the overall information.
DR DZIUBAN: Our experience in New York State, obviously under public scrutiny, is that you address what is probably the single most difficult issue of public reporting, which is the contamination of our previous patient/physician relationship and decision-making process. The concern of the surgeon for his or her own welfare in reporting numbers is, "Am I going to take a hit by doing this case?" And it’s the biggest problem, of course, in salvage cases where the risk is high and the benefit questionable. We keep telling ourselves we have to make our decisions in the patient’s interest, we keep telling ourselves not to worry about the numbers and work it out the best you can, but this is a very difficult problem. I think other factors are going to come into play, namely, economic ones, as we get into issues of resource constraints on this matter.
DR GEORGE C. KAISER (St. Louis, MO): Fred, I would first of all like to congratulate the panel for their outstanding presentations. You have exceeded my expectations. I wanted to address the earlier question about whether the Society should be deciding who should be operated on and who should not. As a physician, when an individual patient comes to you, you have a covenant with that individual to do the best that you can do for that patient. The Society is a collection of individual physicians. I would submit that the decision whether you should operate on a certain patient or not is not the Society’s decision. There is an important aspect of these difficult decisions, however, that needs to be addressed. If extending procedures to those with potentially little benefit uses considerable resources, the decision of whether operation should be offered to them is a societal decision and should not be made by an individual surgeon.
DR GROVER: Thank you for all of your interesting comments. On behalf of The Society, I want to thank you for taking time out to attend this session, and for your many excellent questions and discussion.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
