Beware Knives

In the US, new drugs are not allowed until a randomized clinical trial suggests they are safe and effective. New surgical techniques, however, require no such tests. This isn’t the only bias favoring surgery over other treatments:

In the JAMA study, … researchers found that some women with early stage breast cancer gained no survival benefit from removal of the lymph nodes even though cancer had been found in the lymphatic system. This finding sparked a wave of publicity, including an insightful Room for Debate feature in the New York Times that included 7 authors’ perspectives on whether American surgeons promote unnecessary surgery.

I have no doubt that many of the issues raised by the New York Times commentators are important. Surgeons do have financial incentives, established practices, and natural responses to clinical uncertainty that lead them to suggest surgery in some cases where there is no clinical evidence to support such an action.

Yet, I think we also need to acknowledge that we, the public, also contribute to overuse of surgical procedures. … A few years ago, my colleagues Angela Fagerlin, Peter Ubel, and I published a simple paper titled “Cure me even if it kills me: preferences for invasive cancer treatment.” In it, we showed that people who were presented with hypothetical cancer treatment scenarios tended to choose surgical interventions even when those interventions increased the total risk of death. The effect was much reduced for medication therapies versus surgeries. (more; HT Tony Barrett)

GD Star Rating
Tagged as:
Trackback URL:
  • In large studied groups, the ten year survival rate for men with prostate cancer is not improved by any kind of therapy. Currently, a doctor finishing his urology residency cannot receive board certification in the USA unless he has performed (?participated in) a minimum of 25 radical prostatectomies, and robotic assisted laparoscopic prostatectomy is all the rage. BTW, most men who have these RALP’s and traditional radical prostatectomies, regardless of the claims to the contrary, become impotent, and a significant percentage have urinary incontinence. In general, prior to the treatment, most men who are considered to be good candidates for the surgery have no symptoms at all. Again, their life expectancy, as a group, remains unchanged.

    • Robert Koslover

      “In large studied groups, the ten year survival rate for men with prostate cancer is not improved by any kind of therapy.”

      That is quite a statement. Would you please provide a relevant citation on that? Thanks.

      • I suggest that you do your own research. I know your on-line name from previous comments and as such I am convinced that you have the ability to confirm/deny my assertion. Consider that when this information first was published by the American Cancer Society, the reaction by the American Urologic Association was so strongly negative that the ACS published a revision of their the original opinion stating that the patient should consider his consultation with his private physician as the ultimate decision maker. Start your research with the American Cancer Society.

    • Buck Farmer

      Is there a strong difference in outcomes based on patient age?

      My understanding is that overwhelmingly prostate cancer is an end-of-life phenomenon and usually not deadly (something else kills you first).

      Since impotence and incontinence show up more frequently in elderly men than young men anyway, I’m wondering if the link is causal or merely correlative.

      • Is there a strong difference in outcomes based on patient age?

        Over the past 20+ years, the technique called “nerve-sparing radical prostatectomy” has become prevalent, especially in academic centers. Some non-believer surgeons call it “cancer-sparing radical prostatectomy” because of the tissue left behind. Prior to this, it was a foregone conclusion that men who underwent radical prostatectomy would be unable to achieve adequate erections. Incontinence rates varied wildly. In a recent study from Memorial Sloan-Kettering Cancer Center in New York, the functional outcomes of 1422 radical prostatectomy patients were reviewed with regard to the identity of the surgeon (there were 16 surgeons involved). Adjustments were made in the data to exclude results in patients who were impotent or incontinent pre-op. Other variables were accounted for in matching the groups. The patients of four surgeons achieved less than 65% rate of continence, while those of a different 3 surgeons achieved better than an 80% rate. Others were in between. Regarding potency, preservation rates of 45% were found for 2 surgeons, while 2 others had rates under 25%. The groups were comparable pre-op, the facility was the same, the operation was (nominally) the same. Interestingly, those with the best functional preservation also had the best chemical result when tested for residual cancer one year later. Half of all doctors are in the bottom half.

        My understanding is that overwhelmingly prostate cancer is an end-of-life phenomenon and usually not deadly (something else kills you first).

        Prostate cancer is graded by a “Gleason score” which generally corresponds to how severely mutant the cells are. Higher scores metastasize (spread) earlier. Prostate cancers found in younger men are much more likely to have high scores. Extrapolation from autopsy studies suggests that if every cell of the prostate were to be examined in men over 50, it would be much more likely to find prostate cancer than not. The catch-22 is that the majority of the highly aggressive cancers have already metastasized prior to diagnosis, even though tests cannot detect it; there is no “cure” for metastasized prostate CA. Low Gleason score cancer tends not to be the ultimate cause of death. Putting these two factors together, it is not surprising that overall, the treatment of prostate cancer has no statistical benefit on the survival of the entire cohort of those at risk. Most studies recommending the various treatment modalities concentrate on elimination of detectable disease being a superior result to the detectable disease in untreated patients, rather than concentrating on the survival rates.

        In fact, a recent study from Duke University (Donatucci & McNamara) found that not a single USA Urologic Oncology Fellowship program considers the topic of cancer survivorship to be part of the training program.

    • jor

      Although I agree with your general sentiment that prostate cancer is over-treated, I think your views are not exactly correct. For high risk prostate cancer local therapy (radiation) improves overall survival — c.f. <a href=" SPCG-7 compared to hormonal therapy alone. 10% absoulte benefit.

      Although Bill-Axelson’s study of RP vs WW is no longer significant for OS, it still demonstrates a benefit for distant recurrence and prostate related death —- although the patient population studied most certainly had much more aggressive cancers than the PSA detected cancers seen today.

      I completely agree prostate cancer is over-treated, I just think its important to make sure people realize there is still is a decent subset of patients that NEED LOCAL treatment.

      • I completely agree prostate cancer is over-treated, I just think its important to make sure people realize there is still is a decent subset of patients that NEED LOCAL treatment.

        You quote studies in which participants are highly selected, but fail to consider how these patients came to be diagnosed. In order to find such asymptomatic patients, the general population has to be screened. When that is done, statistics tell us that the overall good achieved is matched by the harm done. In other words, to help the cohort you describe, you must harm an equal number of others. My statements addressed the general cohort of all men who might have prostate cancer.

        It can be difficult to accept such statistical conclusions because they may seem counter-intuitive or even callous. But as Eliezer Yudkowsky once wrote to me: “I don’t think you understand what statistics mean. They are not a sort of weak extra argument that you weigh in addition to your much more reliable personal experience; statistics are a stronger, more reliable way of looking at the world that summarizes far more evidence than your personal experience, even though it just looks like a little number on paper while all that other experience weighs so heavy in your mind.”

  • Buck Farmer

    While I generally agree with Robin that our preferences for medical intervention are driven by considerations other than health…

    …I think that the trade-off between longevity vs. quality-of-life may be a mitigating factor in this particular case.

  • One day when I was a junior medical student, a very important Boston surgeon visited the school and delivered a great treatise on a large number of patients who had undergone successful operations for vascular reconstruction. At the end of the lecture, a young student at the back of the room timidly asked, “Do you have any controls?” Well, the great surgeon drew himself up to his full height, hit the desk, and said, “Do you mean did I not operate on half the patients?” The hall grew very quiet then. The voice at the back of the room very hesitantly replied, “Yes, that’s what I had in mind.” Then the visitor’s fist really came down as he thundered, “Of course not. That would have doomed half of them to their death.” God, it was quiet then, and one could scarcely hear the small voice ask, “Which half?


  • Max M

    I’ve heard that doctors push circumcisions in this very way.

    Why don’t we see private solutions to this problem? A “consumer reports” for surgical procedures, available to the public, would be a great here. I suppose we don’t see this because medicine isn’t really about getting healthy, as you’ve suggested Robin.

  • jor

    Robin — although the general point of testing surgeries rigorously is a valid one — the particular jumping off point — breast cancer surgery is a terrible example. Breast Cancer Surgery is probably one of the MOST tested surgical procedure. Innumerable permutations have been tested in RANDOMIZED trials going back to the early 70’s (maybe even 60’s) — c.f NSABP.

    In terms of testing Axillary Dissection (a component of the surgery) — maybe this trial could have been done 5-10 years earlier, but not really much earlier than that. This is in the context of the history of treatment of breast cancer.

  • Alex Weiner

    I’m wondering how one could ethically perform randomized, double-blind surgeries. What would the placebo be?