Randomized controlled trials - God's gift to medicine???

Hi there.
I saw that there was a link to a letter to the editor of Fertility and Sterility from 2005 by a Dr. Clark in England posted in one of the chat rooms frequented by women with endometriosis. The letter referred to excision of endometriosis and the lack of randomised controlled trials of excision, concluding that unless and until it is proven by these randomised trials to be superior to other therapies, then it should not be recommended as a treatment. There are several issues raised in this letter that I disagree with, and I will try to explain them as simply as possible, although some of the concepts are difficult even for some doctors to understand. But here we go...
First, Dr. Clark is of the mindset that prospective randomised controlled trials (RCTs) are the only way to document effectiveness of a treatment. To understand why he is wrong takes a little background into medical research and the different types of studies. When patients are enrolled prior to treatment and randomly allocated to one of 2 or 3 groups that will undergo different treatments for the same disease, then followed to assess results, this is called a prospective (done in real time) randomized (randomly assigned) controlled (the control group is one group that either has no treatment or a well established treatment) trial. Placebo controlled trials are typically done to see if a given drug has more effectiveness than the placebo effect, which is where people think they're getting therapy so they feel better. Placebo effects are well documented for both surgery and medical treatment, and about 30% of people typically respond to placebo for a limited amount of time. Once a drug has been on the market for a while and we know it works (for example cholesterol lowering drugs), it would be unethical to have a treatment group that gets no therapy (placebo), so one drug is compared with another drug (ie Lipitor vs Vytorin) in a randomized fashion to see which drug lowers total cholesterol and LDL the most. RCTs are typically felt to be the best quality evidence available because of the lack of bias (imposing the author's own perspective on the results). However, RCTs are rarely done to study surgery, and their quality depends on how they were set up. The Women's Health Initiative (WHI) is a perfect example of a huge (15,000 pts) study that was a randomized, blinded, placebo controlled waste of money (your money as it was funded by the NIH). This was the study publicized by CNN and all the network news programs about 4-5 yrs ago where all the TV people said "stop your prempro!" because it causes heart attacks and breast cancer. The goal of the study was to see if hormone replacement therapy actually protects women from heart disease as we thought for many years based on cohort studies (see below). Unfortunately, the study design was flawed and instead of studying women who were going through menopause around age 50, they chose women with an average age of 64 who were in most cases 10 years or more past the menopausal symptoms and off HRT for that long. Then, they started the treatment group women on hormones (prempro), and tabulated the incidence of breast cancer, heart attacks, strokes, and in a sub-analysis, mental function. I don't have time to go into all the details, but suffice it to say they were shocked at the findings, and actually stopped the study early. The women in the treatment group had a higher incidence of heart attacks and dementia, contrary to what had previously been thought. After further review and several more studies designed to look at hormone replacement in women who need it (around age 50 and within 1 year of starting menopause), guess what? The women on estrogen DID have a lower risk of heart attacks AND had better mental function. The multi-million dollar RCT was wrong, and it was all because of a faulty design. Just like the Titanic (well, sorta).
The next type of study we'll examine is called a cohort study. These can be done either prospectively or retrospectively, and involve groups (cohorts) of patients that undergo a particular treatment and subsequent followup. These are the type of studies that are most often done to assess surgical outcomes. Typically, a specific type of patient is selected (ie stage IV endometriosis with no previous surgery) and a specific type of procedure is done to them (ie excision of rectovaginal endometriosis via laparoscopy) and certain outcome measures are followed over a period of time. The more specific the outcomes (quality of life questionnaires, rectal pain, lack of tenderness on exam, etc) and the longer that patients are followed without losses (ie if 100 pts had surgery and 90 are followed for 5 years then 10% are lost to follow-up) then the better the quality of information. Most of the data regarding the surgical treatment of endometriosis is in this class of studies, and most of the studies on excision clearly state the outcomes measured, and also the numbers lost to follow-up. When multiple studies done on different continents by different surgeons all agree with nearly the same results, it is probably the best quality evidence that could possibly exist. This is the case with the excision literature. One study is based on laparotomies (open surgery) with excision done on all stages of endo (Wheeler and Malinak, 1987). Redwine in 1991 wrote about a cohort of women who underwent conservative excision (without hysterectomy or oophorectomy) via laparoscopy, followed for 7 yrs. Redwine and Wright in 2001 studied women with complete obliteration of the culdesac who underwent excision, then followed them for on average 5 years. Abbott and Hawe in 2003 studied women who underwent excision in England and Australia with a 2-5 yr follow-up period. ALL of these studies had nearly identical results - a 20% likelihood of persistent endometriosis over a long follow-up period. To me, this is much better evidence that excision works than any randomized controlled study could ever be.
Another note on the RCT. I don't believe it's ethical to randomize women with pain into a surgical treatment arm vs a surgical placebo arm (making the incisions and not removing the disease). The only ethical way to do a RCT of surgical treatment of endo would be to compare excision and ablation of early stage disease. We shouldn't do it for stage IV disease, because ablation of deep disease doesn't work, as proven by cohort studies. This study of excision vs ablation for mild disease was actually done in 2004, but with only 12 pts in each group followed for 6 months. Not surprisingly, there were no differences in outcomes. Again, it seems that the better quality data comes not from the esteemed RCT, but from the consistency of results in cohort studies which have remained the same over time and continents.
Dr. Clark's second error is that he confuses chronic pelvic pain (CPP) with endometriosis. While it is true that endometriosis is one of the conditions that comprises CPP, it is the only one that is amenable to surgical excision. Several factors need consideration in order to understand the dilemma at hand. The main symptom of endometriosis is pain, but there are many other things that can cause similar pelvic pain. One of the problems we have in studying endo and responses to treatment is that we can't re-operate on everyone to assess whether or not they had resolution of their endometriosis after excision. We re-operate on the patients who have recurrence of their pain, and in general we find that about 60% of them do Not have endometriosis, ie, they're cured. Ironically, it's the patients with the deep disease (typically stage IV, some stage IIs) that have the best chance of pain relief after excision. Why is this? Most people think it's because deep endo almost always causes pain, and superficial disease may be asymptomatic. When a patient with pelvic pain from another cause (such as interstitial cystitis or IBS) undergoes a laparoscopy and is found to have stage I endo and treated for it, the initial assumption is that the pain was from the endo. But maybe the pain was actually from the IC, and the endo was a red herring. You see the problem, I think. This is why it is troubling to follow patients by symptomatology, yet we can't reoperate on every patient 5 years after their initial surgery just to see if the endo is gone, especially if they are asymptomatic.
Without a doubt there is room for more research to be done in order to direct clinicians towards the best way to treat their patients with endometriosis. This doesn't mean, however, that we "throw the baby out with the bathwater" just because we don't have the type of evidence some people want. We have to practice evidence-based medicine, yet there is always the need for common sense in order to interpret what we read.
I hope this is understandable to you all. One thing I've been impressed with since I've been here in Bend is how great an understanding of medicine many of our endo patients have. It's refreshing to have patients who truly understand treatment options, risks, benefits, and alternatives; not just not and say "Yes, doctor".
Anyway, have a great week, and let me know if you have any more questions.
Dr. Mos
ps - for those of you that wrote, next week we'll discuss the finances of medicine.

Almost a year

Hi everybody, hope you're having a great summer.
I realized the other day that it's been almost a year since I left Hawaii on my quest to become the next endometriosis specialist. I've truly learned a lot, more than I could ever put into words, and I've been lucky to have a phenomenal teacher in Dr. Redwine. He's taught me surgical techniques, anatomy, clinical diagnostic pearls, persistence, and the ever important quality of trusting your gut. Working with David has made me realize what a great man he is - he's not just an excellent technical surgeon, but he has that thing that's hard to quantify that makes someone special - the ability to relate to people where they are, to understand their situation, and be able to reach out and help them transform their lives. I am humbled every day, and at the same time I strive to be worthy to follow in his footsteps. I'm growing in my surgical abilities - what once was moderately difficult is now easy, and what seemed nearly impossible when I got here is challenging, but doable. You all know that I've been moved by your stories of frustration over multiple failed surgeries, and that knowing what you've been through both before and after trips to Bend continues to motivate me to "fight the good fight", enlightening the medical community and patients alike about the benefits of surgical excision, the shortcomings of Lupron, and the need to realize that endometriosis can be cured. Last week we read an article by a bunch of PhD microbiologists on the genetics of experimental endometriosis induced in rats ( who don't naturally get endo) and in multiple places in the article they said endometriosis was incurable. I'm realizing that it's now my job to publicly correct these old outdated incorrect opinions thrown around as facts.
So, hang in there, and remember I'm here for you. Any questions you want answered let me know and I'll do my best.
Dr. Mos

First of all, thanks to all you ladies who wrote back about your relationships. I was especially touched by the last comment, and it's so true that endo affects more than just the spousal relationship, but also the ones with your kids, your parents, your friends. Seeing the pain our patients go through before getting any relief is what drives me to excellence, because it's only by my ability to do what Dr. Redwine does that I'll be able to carry on his legacy of helping patients who find no help elsewhere. It's gratifying to know that David thinks I'm almost there - I can successfully excise all endo in about 95% of our patients, and it's only the really really tough cases (who most other endometriosis surgeons have failed) where I need his help. I'm pretty excited about that. But I digress.....
I was asked to write about endo and infertility by Nancy Peterson. I guess she's had a lot of questions about this and thought I might shed some light on this. I told Dr. Redwine about my quest, and he said "it's a black hole"....
So, what can be said about infertility within the context of endometriosis?
First, endo patients typically are slightly less fertile than women without endo. That being said, there are many different etiologies for sub-fertility, and many may be at play in any one couple. Plenty of women with endo conceive and carry perfectly normal pregnancies, so why is it that some can't? There are the obvious reasons such as those with tubal adhesions where the tubes are stuck to the back of the uterus, or to the colon, or anything else, really. But those people are few and far between. Why does mild endo affect fertility? I pulled a couple of articles yesterday that pertain a little to this. For women with mild-minimal endo, the conception rates and delivery rates after IVF were much better in those women who underwent surgical excision of the endo prior to IVF. This suggests that even when bypassing the tubes via IVF, the endo living in the pelvis has a negative impact on fertility. Probably this is mediated by some local inflammation, but the exact reason hasn't been elucidated yet. Treatment with any medicine, whether birth control, lupron, or aromatase inhibitors only decreases fertility, and has never been proven to enhance fertility in patients with endo.
Finally, conservative surgical excision (what we do - remove the disease, not uninvolved innocent organs) has a beneficial effect on fertility. Dr. Redwine wrote a paper about conservative excision for obliteration of the culdesac (stage IV endo), and 40% of the women who wanted to conceive after surgery did. That's pretty good, considering that most of those women hurt too much beforehand to even think of getting pregnant. Surgery is the only treatment for endometriosis that improves fertility, although it is not a guarantee of fertility. The only time fertility is challenged by surgery is when the patients have large bilateral endometriomas, and after removing the cysts, at times there is not a lot of normal ovary left in place, so the number of follicles (potential eggs) is greatly decreased after surgery. Unfortunately, there's no other way of treating endometriomas that works without risk of recurrence, so pretty much that's what we have to do to get pain relief. Luckily, this is a rare situation and occurs mostly in women in their late 30s and 40s. Stimulating ovaries such as with clomid or pergonal (used with IVF cycles) is contraindicated in women with large cystic ovaries, so there's really no other alternative to ovarian cystectomy in patients with large endometriomas.
So, to summarize this rambling stream of consciousness blog, surgery preferably with excision of endometriosis lesions is the best first-line treatment for women with endo and infertility. If pregnancy is not achieved spontaneously after surgery, then success rates with IVF will still be better after surgery than before.
For those of you with endo who haven't tried to get pregnant - don't give up. Plenty of women with endo get pregnant spontaneously, the old fashioned way, without any help. For those of you with pain, excision is the most effective way to permanently get rid of the endo and preserve fertility, as well as natural hormone production.
Take care, and have a great week.
Dr. Mos