David Olive, MD

A Lively Roundtable Discussion on a Controversial Topic*

Whereas the surgical approach has long been the standard of care for 			ectopic pregnancy, a wider range of treatment options is now available--leading to a number of questions for the 			OB/GYN. Our expert panelists debate the relative merits and drawbacks of methotrexate and various invasive procedures, 			and also discuss their preferred diagnostic approaches, in managing the patient who presents with signs and symptoms 			of ectopic pregnancy.

 Today we are going to address the latest diagnostic and treatment options for ectopic pregnancy. 	During the last 2 years, our hospital has used the methotrexate protocol to treat these pregnancies, and the staff 	has found it to be economical and effective. However, it requires reliable laboratory studies and an upgraded radiology 	department. Dr Stovall, what is your experience at the University of Tennessee, Memphis?

 We have treated more than 300 outpatients with intramuscular (IM) methotrexate. Our success  rate is between 93% and 95%. And methotrexate is certainly less expensive than surgery. For example, the total 	cost for diagnosis, treatment, and follow-up is about $1500, as compared with $6000 or $7000 for laparoscopic salpingostomy.

 What kind of long-term follow-up do you have on these 300 patients? How has methotrexate affected 	their ability to conceive?

 Those who conceive do so very quickly, and function much like women with normal fallopian tubes. 	Ninety percent conceive within 3 months, and 80% of those achieve pregnancy. Of those who do achieve pregnancy, 	about 14% have a recurrent ectopic pregnancy.

 Do most of these patients have dilatation-and-curettage procedures (D&Cs) at the time of 	diagnosis?

 No, most do not. Only those with abnormally rising human chorionic gonadotropin (hCG) titers--less 	than 50% in 48 hours in association with an hCG titer of less than 2000 mIU/mL--have a D&C. If the patient's 	hCG titer is above 2000 and it's rising abnormally, then we go straight to methotrexate, assuming that it's not 	an assisted reproductive technology (ART) patient--where the numbers are different.

 In your estimation, what percentage of these cases are ectopic pregnancies versus pending miscarriages 	or nonviable intrauterine pregnancies?

 I don't think any are nonviable intrauterine pregnancies. Those women who have hCG titers of 	less than 2000 undergo a D&C; those who have falling hCG titers either have an abortion or a spontaneously 	resolving ectopic pregnancy, and thus are not treated. If their hCG is above 2000 and they have an empty uterus 	on ultrasound, then the chance of an intrauterine pregnancy is virtually zero.

 In our patient population, if we see an obvious ectopic pregnancy in the patient's fallopian tube 	and the hCG level is less than 2000, we treat rather than do the D&C.

 We do D&Cs for a couple of reasons. First, methotrexate can be potentially toxic. I never 	want to use it on someone who really does not need it. Second, if you do not perform D&Cs on these patients, 	it is very difficult to publish your data in a peer-reviewed journal. Statistically speaking, many patients are 	going to have abnormal intrauterine pregnancies. Your way is very safe and avoids operating room and anesthesia 	costs associated with D&Cs. However, we do suction D&Cs in the office, and, since we are not a freestanding 	surgical center, our only charges are professional fees and supplies.

 Are there any studies using methotrexate that address intrauterine pregnancy rates following 	previous salpingectomy for ectopic pregnancy?

 Yes, we have a whole group of patients who have had previous ectopic pregnancies. Now with the 	second ectopic pregnancy, they only have one tube left. Many of these patients also become pregnant.

 Dr Olive, several years ago, you were a proponent of doing outpatient laparoscopy for chronic 	pelvic pain where you mapped the pain. Would you consider using this as a diagnostic tool for ectopic pregnancy?

 Yes. Part of the reason was alluded to earlier by Dr Stovall, when he talked about the different 	costs or charges associated with a laparoscopic approach versus a medical approach. The fees for a laparoscopic 	procedure in an operating room in an emergency situation at night are substantial. However, the charges and, in 	fact, the actual costs of doing the procedure in an office under local anesthesia are 4 to 8 times lower than those 	incurred in a hospital.

 The charges at Yale-New Haven Hospital run from $5000 to $8000 for a diagnostic laparoscopy. In 	our office it is between $800 and $l700. If you look at the actual costs rather than the charges for doing two 	cases per week in the office--including equipment costs, staff time (from maintenance to professional), and instrument 	depreciation--it is around $800 per procedure. In the outpatient surgery center across the street, it is twice 	as expensive--around $1600 or $1700.

 We've looked at simple laparoscopy done under local anesthesia as a potential tool for trauma 	at Charity Hospital (in New Orleans). Others have thought that this may replace actual diagnostic laparoscopy. 	I think the scope and the optics are limited.

 I don't think there is any problem with the scope or the optics. The new 2-, 3-, and 5-mm scopes 	are all capable of the job. They're all above the necessary threshold to see everything. The problem comes with 	doing these procedures under local anesthesia. Perhaps we could do this with a semiconscious trauma patient, but 	it becomes a technically difficult procedure to accomplish in a patient with an acute abdomen or distended bowel. 	We originally thought that patients presenting to the emergency department with acute abdomens would be ideal candidates, 	but it is not a valuable tool in these situations. We tend to use it in patients complaining of pain whom we have 	treated medically. Before we rush them off to an operative laparoscopy, we do a simple laparoscopy under local 	anesthesia.

 Dr Seifer, do you prefer surgery to medical treatment with methotrexate?

 My concern about single-dose IM methotrexate is that although it is widely used as primary treatment 	for ectopic pregnancy, it has not been well studied. I am only aware of one published trial from the Netherlands 	that randomized patients to multi-dose IM methotrexate or salpingostomy (about 50 patients in each group).2 The 	researchers found no differences in terms of successful treatment, preservation of the fallopian tube, or patency 	of the fallopian tube. Fertility outcomes were not reported. An abstract from the University of Southern California 	described a randomized comparison between salpingostomy and IM methotrexate (fewer than 20 patients in each group).3 	At 3 months, tubal patency rates were 20% with methotrexate versus 85% with salpingostomy; however, pregnancy rates 	did not differ at 6 months (about 20%).

Although single-dose methotrexate has been around since 1991, it has not been studied in an organized, definitive 	manner. When I ask medical students and residents, What is the proper treatment for ectopic pregnancy?' 	their response is 'methotrexate,' even though no clear-cut clinical evidence exists to support their reply.

 Our three priorities are cost, convenience, and pain avoidance. The methotrexate protocol is cheaper 	and more convenient, and causes less pain. The patient experiences a significant amount of pain when undergoing 	laparoscopy as opposed to a simple IM injection. Methotrexate management is a large leap forward, but it does require 	more physician follow-up time.

 All of the comparative data collected so far have been  derived from small, randomized trials. 	Methodology has generally been poor, sample sizes have usually been  inadequate--even when linked together via meta-analysis--and 	the questions have not always been pertinent to the practice of  medicine in the United States. We do not yet have 	the large, definitive, randomized trial comparing major modalities. We  do have a follow-up to the Netherlands study

 described above, which found that methotrexate recipients, as compared with patients who 	underwent laparoscopy, had inferior quality-of-life scores.4 They had more discomfort and a decreased amount of 	physical activity afterwards. In addition, for those patients with an elevated hCG (>1500) but no initial screening 	laparoscopy, as well as for those with an hCG greater than 3000, it was more cost effective to perform laparoscopy 	than to use medical therapy.

 I have never said that methotrexate was better than anything. It is an alternative to surgery 	and costs less in select patients. I do not know whether it is cost effective in terms of quality of life. My impression 	is that it is probably equal to laparoscopic surgery but more cost effective. Unfortunately, no large, prospective, 	randomized studies have been conducted to confirm this impression. Until such time, I think we are going to have 	to settle for the fact that methotrexate works well and costs less.

 I think that we are missing a major point when we talk about laparoscopic salpingostomy versus 	methotrexate use, because most people out there probably still do laparotomies. What is the role of laparotomy 	in ectopic pregnancy, and are we doing way too many of them?

 I find it interesting that when our residents, who rotate through various sites, are in the academic 	center they do laparoscopy. But if it is up to them, they prefer to do a quick laparotomy in about 15 minutes rather 	than use a scope, which takes about 45 minutes to an hour.

 In fact, the data suggest that laparotomy and laparoscopy are probably equivalent in certain 	people's hands. Total costs probably do not differ much. The persistence rate with laparotomy is probably less 	than that with laparoscopy. No one has ever really addressed the failure rate of laparoscopy, meaning that you 	start out doing a salpingostomy, but you end up with a salpingectomy because you have bleeding. I think the decision 	depends on the doctor and the patient. Methotrexate is an alternative, not a replacement, to surgery, and it is 	complementary to surgery.

 I want to ask about hCG levels. Are you finding a methotrexate failure rate when you reach certain 	hCG levels in your institution? That is, do you have a lot more failures in patients with a level of 5000 and more 	successes in those at 2000? Have you looked at these data?

 A lot of people talk about methotrexate failure rates being higher at certain progesterone values, 	when hCG levels are above 1000, 1500, or 3000. Our case series, which is the largest to date, indicates no difference 	in failure rate based on hCG titer, progesterone titer, human chorionic somatomammotropin (a placental marker) 	titer, Doppler flow characteristics, or ectopic size (up to 3 cm).5 The bottom line is that I can't predict who 	is going to fail with methotrexate or have a ruptured ectopic pregnancy, and who is going to respond to methotrexate.

My gut feeling is that, even with methotrexate, we probably overtreat; most of the women on whom we operate 	or to whom we give methotrexate would probably have spontaneous resolutions if we were patient enough to wait. 	Unfortunately, we cannot predict who those people are, so waiting is a dangerous game.

 We need to clarify what we mean when we talk about failure with methotrexate. If we have a failure 	with salpingostomy, it is clear that the patient has to have another treatment, whether it be surgery or methotrexate. 	But when we talk about methotrexate, the failure might be that the patient has had only a single dose and needs 	another dose a week later. To others, the failure might mean that the patient needs surgery or has a rupture of 	the fallopian tube.

 When I say methotrexate failure, I mean someone who needs surgery to have her tube removed because 	of rupture. About 4% of our patients require a second dose of methotrexate, which is simply a part of their treatment 	course.

 The problem with the data is that different physicians have different thresholds for the start 	of surgery. Some wait much longer than others do.

 I agree. If you are not willing to wait, and if you are not willing to play by the rules that 	are currently published in the literature, then you probably should not be using methotrexate; your failure rate 	is going to be so high that you are better off operating to begin with. Certainly, from a cost-effectiveness standpoint, 	it makes no sense to give everybody methotrexate and then operate 2 days later. You must establish a threshold 	so that you don't operate on more than 5% to 6% of the true failures.

 A study conducted at the University of Southern California randomized patients to salpingostomy 	alone or salpingostomy plus one dose of IM methotrexate. The persistent ectopic rate was reduced from 15% to less 	than 2% with the addition of methotrexate.6 The incidence of side effects was about 5%. In a patient population 	where you have to worry about compliance, this may be something worth considering. Otherwise, you end up doing 	a laparoscopy, cannot find the ectopic, and then do not feel comfortable with whether or not you removed the entire 	products of conception. So you follow up with methotrexate within 24 hours.

Dr Stovall, since 40% to 60% of patients have pain within 2 to 7 days after methotrexate, can you provide us 	with guidelines about which patients seem to do well, do not require follow-up treatment, and are having pain because 	of hemorrhage in the fallopian tube that has no clinical significance, versus those who are really experiencing 	a rupture?

 Actually, I cannot give you any guidelines. Many doctors will immediately take their patients 	into the operating room as soon as they complain about pain, even though most of them have not experienced a rupture. 	Our guideline is that we give patients some sort of pain medication--Tylenol¨ #3, Motrin¨, Advil¨--and 	ask them to call if the medication does not work. If the pain is severe and persistent, we bring them back to recheck 	their hemoglobin and hematocrit values to ensure that they are stable.

Frequently, the patient has already had an ultrasound. And ultrasound data suggest that if you follow patients 	treated with medical therapy with ultrasound, their ectopics will get larger, they will have bleeding into the 	ectopic, and they will have increased blood in the cul de sac. If you are going to take all of those people to 	the operating room, you are probably better off skipping the methotrexate. We take only those patients who have 	a fall in their hemoglobin and hematocrit values along with pain, as well as evidence of a ruptured ectopic pregnancy, 	to the operating room.

 Can we talk about ultrasound and discriminatory zones? Does anyone have problems with your laboratories 	or with changing assays? How important is it to know your lab--especially the discriminatory zones? My concern 	is that the accuracy of the hCG level varies depending on the lab. Let us start with Dr Olive, who is at Yale, 	where discriminatory zones were first described in 1981.

 I think that it is critically important to know your laboratory--including the assay and the International 	Reference Preparation they are using--and how these findings are interpreted in terms of discriminatory zone. But 	let us not overlook the potential problem that exists with ultrasound. We assume that ultrasonography is going 	to adequately pick up intrauterine gestation if the hCG level is over a certain threshold. That may or may not 	be true, depending upon the type of machine and the ultrasonographer. At Yale, we have a major problem in the evening 	when we do not have access to an ultrasonographer. We send the studies to radiology, where we get at least as much 	variation in the interpretation of ultrasounds as we do in hCG levels from the lab.

 Some members of the emergency medicine community have proposed that they should be the ones doing 	the ultrasonography. In a paper we published several years ago comparing the skills of house staff versus staff 	physicians in interpreting abdominal ultrasounds, we found that accuracy was contingent on experience level.8 We 	recently had a case in which an intrauterine pregnancy was missed. A quality assurance review revealed that the 	ultrasonographer and technician had not gotten a full-length view of the obese patient's uterus, and the pregnancy 	was high in the fundus. It was later discovered by follow-up ultrasonography.

To summarize, it is critical for clinicians to know two major variables--the laboratory they are dealing with 	and the quality of their own ultrasonography.

 We never rely on a single hCG value. And we never treat a patient in the emergency department, 	even if she has an hCG level of 10,000 and no intrauterine pregnancy, based on a resident's or emergency department 	physician's ultrasound interpretation. The patient always comes back for a follow-up hCG titer to make sure that 	it is indeed rising, and then she has transvaginal ultrasound done by someone with experience. We always go the 	extra mile to prove that it is truly an ectopic pregnancy.

 And perhaps the other caveat is that we each have different patient populations. If
 	we are dealing with patients managed by ART, who may have heterotopic pregnancy rates of as high as 2%,9 you can 	get yourself in a difficult situation where you are using methotrexate to treat the ectopic pregnancy and you end 	up compromising the intrauterine pregnancy.

 If we see a patient who has a very rapidly rising hCG titer and it is 2000, and we do not see 	any intrauterine pregnancy, we wait another 48 hours to make sure that she does not have a twin pregnancy or a 	heterotopic pregnancy. If she has had ART, then the rules really change. The good news, though, is that you know 	the exact moment in time when those patients become pregnant, so it is much easier to catch them early on.

Articles

Whereas the surgical approach has long been the standard of care for ectopic pregnancy, a wider range of treatment options is now available--leading to a number of questions for the OB/GYN. Our expert panelists debate the relative merits and drawbacks of methotrexate and various invasive procedures, and also discuss their preferred diagnostic approaches, in managing the patient who presents with signs and symptoms of ectopic pregnancy.

Update on Ectopic Pregnancy: Diagnosis and Treatment

click             here for RealAudio/Video version

       "My name is Dr. David Olive, and I'm at Yale University School of       Medicine."
       
       

       "I'm Dr. Bill Schlaff, and I'm at the University of Colorado."
       
       

       "We're involved in a conference regarding endometriosis and a number       of controversies regarding the treatment of endometriosis. What we thought       we would do is discuss three of those in particular. The first being the       patient who fails some type of treatment plan for endometriosis associated       pain or at least pain that's presumed to be from endometriosis. They've       been treated with oral contraceptives or with non-steroidal       anti-inflammatory drugs, and then the question is - how would you proceed       from that point? Do you proceed with surgical intervention with a       laparoscopy to make the diagnosis and then treat or should you presume to       just know that the patient has endometriosis and treat with GnRH agonists       in a presumptive way? Bill, I'd like to ask you how you would proceed at       that point?"
       
       

       "What I would do is really very consistent with what our consensus       showed in our conference today. We had upwards of three hundred physicians       from all over the world, predominantly northern and southern Europe, but       some from the states and a variety of other areas throughout the world.       The conclusion really in all of them - the majority I should say - was       that the next step would be laparoscopy rather than a pair of treatment.       The vast majority would treat or perform surgical extirpation of the       endometriosis, generally speaking, with electrocautery at the same time.       That's actually what I would do as well. In the U.K., it's quite       interesting because of the management of operative lists and the like.       They would often simply do a laparoscopy, make a diagnosis, and then       subsequently treat with some type of medical therapy or bring the patient       back for a treatment laparoscopy."
       
       

 "I       would agree that the predominant approach, in the world today, is to do       laparoscopy at that point. But I think there's a growing movement that       says you don't really need to make the diagnosis of endometriosis if you       have good evidence that endometriosis probably exists, the patient has       pain consistent with it, and you've ruled out other diseases. I think       there's a growing number of physicians who are comfortable in that       situation using a GnRH agonists as presumptive therapy and seeing if the       patient responds. If the patient does indeed respond, you can simply       continue the medical therapy. If they don't, then you can proceed to       surgery or some other type of evaluation to try and pin down exactly what       might be causing the pain."
       
       

       "I actually extend that even further in many ways because I'd submit       that it is often the case that we really don't know whether pain       experienced by a woman is endometriosis or whether or not it may or may       not be classical. I would suggest, perhaps somewhat cynically, that many       practitioners don't even bother taking a very extensive history but simply       in the face of reported pain that is unresponsive to these other methods -       that David was describing - would move directly to the use of GnRH       analogue independent of whether the consolation of symptoms was even       suggestive of endometriosis. Thereby effectively saying that we are not       going to be concerned about diagnosis in the least but rather we're going       to use an empiric treatment approach to the treatment of pelvic pain of       whatever variety. And only in those women who failed to respond or who       have a rapid recurrence we'll be concerned about making a more definitive       diagnosis by laparoscopy, and I really think that that speaks to the more       difficult problem. I guess what I'd ask you is - in the face of a somewhat       cynical approach or even nihilistic approach to diagnosing it or treating       pelvic pain, while this is interesting in the women who's twenty-six years       old who's life could have a chronic process - how do you fashion a medical       approach or any combined approach to treating her pain, not just for the       next six months but for the next six or sixteen years?"
       
       

 "I       think the real key in trying to figure out how to deal with endometriosis       is coming up with not just a short-term but rather a long-term approach.       What you have to end up doing is using all the tools in your armamentarium       that are available to you. That includes surgery, medical therapy, and       re-treatment with medical therapy - things that have not been well       demonstrated to be effective in good studies in the literature but which       we believe to be effective, and are frequently utilized today. So I think       the idea that you can simply treat, send the patient home, and be done       with it - is one that we now know is not a true picture of the situation.       What we will end up doing is designing plans to last for one, two, five,       ten, or fifteen years as best we can."
       
       

       "That really brings up the whole concept of add-back therapy. A       concept that was first published in the late eighties used progestins       primarily as adding back to an analogue. The theory here was that by using       some sort of an add-back therapy, one could forestall the metabolic and       symptomatic side effects of hypoestrogenism at the same time as one       allowed continuation of the therapeutic benefit of the treatments in a       different way, while the metabolic consequences and side effects were       reduced. There was no attenuation of therapy of benefit. So progestins       were initially used, followed by hormone replacement therapy, more       traditional estrogen plus progestin, and now in a variety of other       compounds, and other approaches including disphosphonates, and tibolone       outside of the United States. PTH has been used, and there may well be       others as well. One thing I'd point out is that throughout all the       published studies, one of the really unmeasured components relates to the       impact of the ongoing interaction between the subject or the patient and       the study site, which is to say - the physicians, nurses, and coordinators       who participate with them. One can't really underestimate the importance       of an interested ear, the either frankly psychotherapeutic or informally       psychotherapeutic approach. One of the things we really have to do is to       better develop our understanding of the specific nature of pain related to       endometriosis versus the non-specific. And begin to look at other       potential ways that we can improve outcomes - things like looking at       evidence of depression and treating that or some would suggest, looking at       nutritional aspects and the like. I think these are all important and,       perhaps, areas we haven't looked at enough."
       
       

       "Yes, I think there's no question that the therapeutic approach that       we use with frequent visits for the pelvic pain patient or any type of       pain patient ultimately becomes beneficial. We also have other tools that       are further beneficial, and we've proven these in the placebo controlled       camp surgery and controlled trials. We know that add-back therapy is       effective, and in fact, in all the studies that we've looked at - it's       been not only effective but also probably enhances compliance by the       ability to remove the side effects. So for that reason, I think both Bill       and I are strong advocates of the use of add-back therapy with GnRH       agonists and the ability to utilize then the therapy for a long period of       time. We noted today in our session that most of the surgeons are       interested in operative laparoscopy at the time of initial intervention       and diagnosis of the endometriosis, and then the question arose - what       should happen at that point? Do we just let the operative laparoscopy       stand and see how the patient does or do we continue with medical therapy       afterwards? Interestingly enough, an overwhelming number of the physicians       from around the world elected to follow their initial operative       laparoscopy with additional medical therapy. Bill, I wonder if you could       comment on that briefly?"
       
       

       "I think that there is first and foremost, a huge gap in our       information. The information gap that I'm alluding to is better knowledge       of the percentage of patients who respond to initial surgical therapy and       the characteristics of return of symptoms. In most studies, with the rare       exceptions absent the control group, and with rare exceptions simply       reporting results as opposed to performing any statistical analysis,       taking into account drop-out such as a live table analysis, most of these       studies will show us a benefit rate of 50%-80% over an indeterminate       amount of time ranging from two or three months up to a year or so. So we       don't really have a good grasp of, first of all, the percentage of       patients who respond by improving their pain, assuming again that this is       endometriosis related pain. Nor do we know the characteristic recurrence       rate of pain over time. Absent that, I think, it's quite difficult to       decide whether or not immediate adjunctive care is important. For example,       if we found that 80% of patients had a two or three year cessation of pain       following surgical treatment, I would be inclined not to want to treat       with adjunctive therapy because presumably 80% of the patients wouldn't       even need it. The other question I think is important, David, maybe you       could comment about, is what the real benefit of adjunctive therapy is.       Said a different way, is it simply a way of providing pain relief for the       six months immediately following surgery or does that six month treatment       produce more prolonged benefit than simply the surgical approach?"
       
       

       "That's an interesting question. We have very limited data to address       that but it looks as if you don't get much more than the limited amount of       time that the patients are on the medical therapy, by which to say you're       extending the relief from the pain. So if you have them on for six months       postoperatively - you get maybe six more months of pain free interval. If       you have them on for a year or two years - you get that much more pain       free interval. So I think that's probably what we're looking at although       again, the studies have been fairly rare and fairly small to this       point."
       
       

       "I would submit that to me that would imply in the presence of       imperfect data, if at all, and the virtual absence of good data said a       different way, that perhaps the best way of approaching this is to allow       the surgery to work or not. If indeed it does, one can hold back the use       of a GnRH analogue for a subsequent time when there's a recurrence. In       fact, when you think about it, we have virtually no data about the       metabolic or symptomatic benefit of re-treatment. To my knowledge, there       is one study not very well characterizing these things. So if that's the       case, if you treat immediately and there is a recurrence which is forced       out by six months or however long, the treatment is what then is the next       step. Do you re-treat, and if so, what evidence do we have that that's       neither effective nor metabolically appropriate? So my view, frankly,       would be in contradiction to the majority of people at our symposium. I       would prefer to treat - simply with surgical extirpation followed, if you       will, by treatment with medical approaches in the circumstance where       there's reoccurrence of discomfort."
       
       

       "If I could sum up real briefly, we entered our session today because       we did determine there was a controversy regarding the treatment of       endometriosis associated pain. We were looking to see if we might find a       consensus, and we found none. I think that's pretty much been the       experience in every place that's looked for a consensus, whether it was       the Canadian consensus report or the United States consensus meeting or in       this meeting today. What we found was a wide disparity in the approach to       endometriosis, and that's not entirely the physician's fault. We have a       few randomized trials, a few studies that address specific questions       regarding the treatment of endometriosis associated pain but there are a       lot of other holes in the literature that have yet to be addressed       properly by appropriate studies. So what we clearly need to do is identify       those questions, design, and carry out large multi-center trials that       address these very specific points so that we might practice in a more       scientific manner."

OBGYN.net Conference CoverageFrom First Congress on Controversies in Obstetrics, Gynecology & Infertility Prague CZECH REPUBLIC - October, 1999

David Olive, MD

Articles

Update on Ectopic Pregnancy: Diagnosis and Treatment

Controversies in the Treatment of Pain with Endometriosis