Discussion

Article Outline

• Copyright

Dr Anil Hingorani (Brooklyn, NY). You guys did a great job. You guys had multiple variables that you were looking at and it made it actually much more interesting when you see that so many of the variables fell out of your analysis except for the one that you mentioned, the case volume.

Why did you pick 20 cases per year? And does it make a difference if you are doing 15 or 25? If you use 20 as a guideline, where did you come up with that number? And did it make a difference if you used 15, 18, 20, 25, if you stratified it? Is it even better to have 30 cases per year, or have a surgeon who is having 30 cases per year?

If the operative times between the high-volume surgeons and the low-volume surgeons were similar, the blood losses was similar, the supraceliac clamping was similar, what was the difference besides the number of cases that you've done over the last year or two? Were they sewing differently? I mean, you mentioned that they had longer transport times in the patients who had low-volume surgeons, but I do not see how that would necessarily result in the results.

Dr Cho. With respect to why we chose 20, the studies that have evaluated this aspect, meaning the surgeon's volume and hospital volume, have a significant discrepancy in terms of their definition of volume. Low volume was defined as those who performed anywhere from 1 to 26 and high volume anywhere from 10 to 26.

And a study from Ontario, Canada, has shown that once surgeons' annual volume met 20 or greater the surgeons' volume did not have any impact on outcome. Another study by Luft (Luft HS, et al. The volume-outcome relationship: practice-makes-perfect or selective-referral patterns? Health Serv Res 1987;22:157-82.) defined high volume surgeon as those who performed between 15 and 26. So, we chose a halfway in between which is 20. It was arbitrary, but there was some reason to it.

In regards to your second question about why there is no difference, we were somewhat frustrated by the fact that there was no real tangible difference between the two groups of surgeons. But one can postulate that in terms of blood product usage, it was noted, although there was no significant difference, patients who were operated by high-volume surgeons tended to have free rupture. And obviously those patients will have a much higher blood loss. As soon as the abdomen is entered, there is a lot of blood that is already accumulated that would translate into blood loss, which would then necessitate higher blood product and fluid administration, while actual intraoperative blood loss may be lower compared with patients who presented with contained rupture. So although there may have been some differences, it may not have translated into any statistical significance in the model that we used.

Also, this study does not really analyze the differences in intraoperative conduct as to the difficulty of the operation or subtle nuances, such as tearing the aortic neck, or whatever that may lead to more operative time or additional procedures. So there may be subtle differences that were not detected by our statistical model.

And the third thing is the number of patients who were operated by high-volume surgeons was only 30% and that number may have been too small to detect any meaningful differences between the two groups.

Dr Keith D. Calligaro (Philadelphia, Pa). Did you try to correlate the volume of ruptured aneurysm repairs with the years of surgeon experience after fellowship? In other words, were the low-volume surgeons the younger attendings in your group? Therefore, are you really correlating volume or experience with the results that you obtained?

Dr Cho. That's a great question. We actually looked at the number of years of experience in our surgeons and that did not translate into any difference. There was a longitudinal study that was conducted in the state of New York by Luft et al, and it showed that actually when they followed the surgeons over several years of time, really very few surgeons actually increased their aortic volume from the beginning to the latter part of the study period. And surgeons who had an excellent outcome at the end of the study also had superior outcome even in the earlier phase of study when their volume was low.

So surgeons' aortic volume does not change a lot over course of time and good surgeons with good outcomes were noted to have good outcome even in their earlier phase of career. I hope that answers your question.

Dr Marat Goldenberg (Reading, Pa). You identified two variables: preoperative shock and free rupture, although I did not see in your results whether they were independent predictors of a poor outcome. If the patient had abdomen full of blood or had to have CPR predict a poor outcome?

Dr Cho. No. The short answer is no. We looked at that, and although there was a tendency toward patients with chronic CPR in the low-volume surgeon group, that did not translate into any significance, as well as the other factors that you mentioned.

Dr Linda Harris (Buffalo, NY). Are the low-volume surgeons at low-volume hospitals and high volume surgeons at high-volume hospitals where the problem may not be the surgeon but the perioperative care, or are they both in the same institutions?

Dr Cho. This study was conducted at the University of Pittsburgh Medical Center. And most, the majority of our cases, were done at the Presbyterian University Hospital. There were several cases that were conducted at the other teaching hospital which is Shadyside. So in terms of the effect of hospital volume, that was not the intent of our study to study that.

Dr Edward Y. Woo (Philadelphia, Pa). Did patients have preoperative imaging and did that affect their outcome? Was the approach, transperitoneal or retroperitoneal? Since 80% of the patients were transferred, did the transferred patients actually do worse because there was a prolonged period of rupture state?

Dr Cho. With respect to imaging, 84% of our patients were transferred in from other referring hospitals. But since these patients were transferred in with the diagnosis of ruptured aneurysms, nearly all of these patients had preoperative imaging. So, very few patients had to have an imaging done after arrival to our hospital system.

The second question about approach … . all but two patients were approached through a midline. Only two patients had a thoracoabdominal approach for a suprarenal aneurysm in one patient, and the other patient had a prior aortic graft implantation and a thoracoabdominal approach was used.

Your third question was the time. When we looked at the time taken from emergency room to the operating room, this also did not translate into any morbidities or mortalities. But there have been studies that show that a delay in the operating room up to 2 hours does not really translate into any morbidities. This has been shown many times in the literature. And one study has actually shown that a delay up to 6 hours does not translate into a significant morbidity. This cannot be applied to everybody. I am sure there are people who die within 2 hours of presentation. And several people have died in our system before they reached the operating room. But in general, time taken to obtain additional imaging would not significantly delay prompt treatment and evaluation of these patients.