Q. What percentage of patients with endometriosis are infertile?
A. This question is difficult to answer since we do not really have exact data on the subject. However, different studies have shown that between 30 and 50 percent of patients who suffer from endometriosis have some degree of reproductive failure. This figure is almost three times higher than the number of women in the general population who have difficulty getting pregnant. In terms of what percent of infertile patients have been found to have endometriosis, again, different studies have shown different figures, but the general consensus is around 15 to 20 percent of infertile patients are suffering to some degree from endometriosis.

Q. In infertile patients with endometriosis, what is causing the infertility? A. At the present time, there is not one single factor that can explain the cause of infertility in all patients with endometriosis. Obviously, the cause of infertility in women with severe endometriosis is much easier to understand. In these patients, the extent of the pelvic adhesions, the distortion of the tubes and ovaries, and occasionally because of the destruction of the ovarian tissue, the infertility can be easily explained due to the anatomical distortions that affect the pickup of the egg by the tube and the effect of severe endometriosis of the ovary and ovulation.
However, the cause or relationship of endometriosis in the minimal stage on infertility is not as clear, and what makes matters more fascinating is that we see many women with endometriosis who have normal fertility. However, the fact that 60 to 70 percent of these patients with minimal to mild endometriosis who also have infertility will conceive after treatment, and the fact that the majority of them will conceive within the first year after treatment, is convincing evidence that the presence of endometriosis can interfere with fertility.
In a study which was done in a group of patients who received donor semen insemination as a treatment for infertility, it was found that the presence of mild endometriosis decreased their chance of pregnancy. So there is evidence that even minimal endometriosis can interfere with fertility.
The mechanism by which endometriosis could interfere with reproduction has been suggested to be:
1. interference with normal ovulation by interfering with development of follicular growth and function;
2. interference with sperm mobility;
3. interference with the release of eggs, thereby causing inadequate or even no ovulation;
4. changes in the peritoneal fluid, thereby producing a hostile environment for sperm and ovum.

Q. Can you explain why patients with endometriosis have adhesions?
A. Although the exact cause is still a mystery, it is the general consensus that adhesions result from irritation in the pelvic areas, and possibly from a secondary inflammation due to a miniature menstruation within the implants in the pelvic area.

Q. I read that endometriosis can cause infertility through a substance called prostaglandin. What is this and is it true?
A. Prostaglandin is a chemical substance which has multiple functions. It was first discovered in the seminal fluid, but later on, it was found in many other fluids in the pelvic and abdominal cavity. Some studies have indicated that the level of prostaglandin in this fluid is increased in patients with endometriosis. This increment has been postulated to be the cause of some ovarian dysfunction in patients with endometriosis, and some of the problems that these patients have with tubal function could also be attributed to prostaglandin. Prostaglandins have also been accused of interfering with sperm-ovum interaction, embryo growth, interfering with sperm motility, and interfering with the function of some central nervous system areas that are responsible for control of reproduction. One recent finding shows increased production of prostaglandin in minimal and fresh endometrial lesions. This can explain why some patients with minimal endometriosis can have significant symptoms (i.e., pain and infertility) despite their minimal stage of endometriosis.

Q. After I was diagnosed as having endometriosis, I was told that I have inadequate ovulation. How common is this?
A. Abnormalities of ovarian function are a common finding in patients with endometriosis. Several theories have been proposed to explain inadequate ovulation and in some patients even lack of ovulation. The fascinating point is that endometriosis is a disease of women with ovulation (by this I mean the patient has to have ovulatory cycles in order to develop endometriosis). However, after the development of endometriosis, the presence of endometrial tissue in the pelvic and peritoneal cavities will cause some degree of abnormality or inadequacy of ovulation. Some studies say that up to 50 percent of patients with endometriosis have some degree of ovulatory dysfunction and this should certainly be taken into account in the treatment of patients with endometriosis and infertility.

Q. I was told that I have luteal phase defect. What is this and how common is it?
A. The luteal phase is the second half of the menstrual cycle. The menstrual cycle traditionally has been divided into two parts. The first two weeks have been called the proliferative phase. This is the time that the ovarian follicle is being developed. The lining inside the uterus is proliferating and getting ready for ovulation and possible implantation.
The second half of the menstrual cycle, or the last two weeks, is characterized by a dramatic increase in the hormone progesterone, which is produced by the part of the ovary which is called the corpus luteum. The presence of progesterone is very critical for proper implantation of the fertilized egg and adequate amounts of this hormone are necessary for implantation. In patients with endometriosis, several studies have indicated that there is a higher incidence of this luteal phase deficiency or luteal phase defect. Another problem which has recently been described in patients with endometriosis is the so-called luteinized unruptured follicle syndrome. In this case, the patient will ovulate but the egg does not come out of the ovary and it becomes entrapped in the ovary, which can obviously account for infertility in some of the patients with endometriosis.

Q. What are my options for relieving my pain?
A. Since the pain in women with endometriosis has been attributed to the presence of inflammatory process as a result of the high levels of prostonoids in the endometrial implants, the use of nonsteroidal anti-inflammatory drug (NSAID) that interferes with the production of these chemicals makes quiescence in the patient. Studies have shown that the use of medication such as naproxen sodium (Anaprox-Syntex) can cause substantial pain relief (in over 80 percent of cases) in patients with laparoscopic documented endometriosis. The same results have also been achieved with ibuprofen. I should mention that these medications seem to be more effective when initiated at the first signs of discomfort rather than waiting for severe pain to develop.
The use of codeine-based pain medications, so long as there is close supervision and limited course (a few days each month) has little addiction possibilities and, in many patients, could be extremely helpful in pain relief.

Q. Do you recommend hormonal treatment for controlling pain?
A. During the 1960s and 1970s, oral contraceptives were used with reported 80 percent efficacy. However, because of annoying side effects, this is no longer an acceptable option. During the 1980s, danazol was a popular and excellent choice for prolonged relief of pain; relief even continued in two-thirds of patients after finishing the treatment course. However, this medication was not free of unwanted side effects either; these included weight gain, swelling, decreased breast size, acne, increased hair growth and deepening of the voice.
In the past decade and with the approval of the FDA in the past five years, a hormone which is produced by the brain has been used for this purpose, GnRh-agonist.

Q. I was told that I had a slight decrease in my bone density while on GnRh-A. Can you explain this?
A. Studies have shown that there could be a three to six percent decrease in bone density after six months of therapy. However, the follow-up at one year shows that in 75 percent of patients, normal bone density was present. The physiological reason for this bone loss is very clear and is due to the decrease of estrogen, the same mechanism that works with women in their post-menopausal years. Clinical research studies have been done and are on the way to find a method of supplementation with low dose estrogen and/or progestin to allow stabilization of bone without reduction of the therapeutic effect. Calcium supplementation has also been recommended during medical therapy. It should be emphasized that routine bone density studies prior to or during the medical therapy are neither recommended nor cost effective.

Q. I took medication for six months, and despite this, my symptoms continued. How can this be explained?
A. With any form of medical therapy, up to one-third of the patients will not respond and obviously will not experience any improvement in their pain or other symptoms. This may be due to the loss or alteration of estrogen receptors and/or enzymatic systems which are necessary for the metabolizing the hormones, or simply inadequate levels of drug delivery to the endometriotic area due to extensive scarring or lack of blood supply.

Q. When do you recommend medical therapy in patients who want to get pregnant?
A. As has been stated many times over, no medical treatment for endometriosis should be initiated before a confirmed diagnosis by laparoscopy. My personal philosophy is that any lesion that can be treated at the time of laparoscopy (by any means, laser, cauthery, surgical removal) must be treated. In cases where I feel I removed all visible lesions, I encourage the patient to try for pregnancy for six to nine months; if pregnancy does not occur, medical therapy might be initiated. However, in cases where, at the time of surgery, we are not able to remove the major bulk of the lesions, medical therapy can be initiated immediately following the surgery. In cases where there is extensive endometriosis and large endometrioma (especially those confirmed by a previous laparoscopy), a pre-operative, three to six month medical therapy can be helpful and may increase the chance of more complete removal during conservative surgery.

Q. I continued to have menstrual bleeding while on medical therapy. Does this mean that the medication was not working?
A. Although it is expected that medical therapy should somehow affect the normal menstrual cycle, not all patients will stop menstruating and many continue cyclic bleeding. In some reports, spotting during therapy has been reported in up to 60 percent of the patients. It is believed that the bleeding is not always a good indicator of sufficient levels of medication.

Q. How do you monitor the medical treatment efficacy?
A. The best way to monitor therapy is through measuring the serum estrogen, which in most patients will drop adequately within one to two weeks. In cases with large endometriosis, a pelvic sonogram every three months can show the decrease in size. Pelvic examination is another suggested testing method; however, serum CA-125 measurements have not been found to be helpful.

Q. How do you evaluate the results of therapy after treatment for endometriosis?
A. Most of us use the patient's symptom relief and improvement in pelvic examination for monitoring the effects of medical therapy. In our experience, the decrease in pelvic pain with menstruation and intercourse as well as the softening of the nodules on the uterosacral ligament and decrease in thickness, hardness induration in the space behind the uterus, in the cul-de-sac correlates with the laparoscopic findings of improvement of endometriosis.
However, the most direct and reliable way to evaluate the effect of treatment is by comparing the laparoscopic findings before treatment with laparoscopic findings after treatment. Obviously, if the aim of treatment is pregnancy, the achievement of pregnancy is undoubtedly a very remarkable way to evaluate the effectiveness of the treatment.

Q. How long do you use danazol?
A. Although the initial clinical investigations were based on six months of danazol therapy, the duration of the treatment should be individualized to the specific needs of the patient. For example, the course of treatment could vary from three months, in the case of conservative therapy, to many years. Certainly it has been used in many studies up to two or three years in patients who are suffering from severe pelvic pain or in patients who do not want any kind of surgical therapy. For achieving pregnancy, usually six to nine months of a high dose (800 mg daily) of danazol has proven sufficient to give adequate results.

Q. Can you get pregnant while taking danazol? If so, does it have any ill effects on the fetus?
A. In high doses (800 mg daily), danazol is a very strong and effective contraceptive. The incidence of ovulation has been estimated to be less than one percent with this dose. However, in lower doses (less than 400 mg daily), the incidence of ovulation is quite substantial; this is why we recommend a barrier contraceptive (such as a diaphragm or condom) to be used by patients taking this dosage. Another factor that increases the risk of pregnancy in patients who are taking danazol is the patient's own neglect in taking medication regularly. It has been reported that patients who take the medication intermittently have a high incidence of pregnancy.
In answer to the second part of the question, the answer is yes. Danazol has some weak androgenic (male hormone) effect, which can produce female pseudohermaphroditism in the female fetus of these mothers taking danazol. Again, this is why we recommend that patients who are using danazol (especially in dosage of less than 400 mg daily) use a barrier contraceptive, and we recommend that they do sporadic pregnancy tests to ensure that they are not pregnant.

Q. How effective is danazol?
A. The effectiveness of danazol depends on the patient's symptomatology and also on the stage of the endometriosis. For relieving pain, the overall literature reports somewhere between 70 and 90 percent of patients with endometriosis will have some relief of pain after danazol therapy. The post-danazol pregnancy rate again depends on the stage of endometriosis. However, the overall pregnancy rate in a group of patients with all stages of endometriosis has been around 40 percent, which is not much higher than in patients without danazol therapy. That is why we believe the laparoscopic laser treatment for mild to moderate forms of endometriosis at the time of initial laparoscopic evaluation is currently the most effective single type of treatment.
Another area in which danazol has been helpful is in the treatment of endometriosis outside of the pelvis; for example, in the lung, bowel and other areas where surgery cannot always be done safely. However, it should be noted that although the symptoms will regress during the course of treatment, regrettably, the symptoms will recur after discontinuation of treatment. This recurrence is one of the major problems we have had in treating endometriosis with danazol, even with pelvic endometriosis. That is why, after danazol treatment in patients with large ovarian endometrioma, if the cyst still persists and the patient is interested in her future fertility, ovarian cystectomy or removal and excision of endometriosis via CO2 laser, laparoscopy or laparotomy is indicated.
At any rate, it has been shown that even after a complete and aggressive course of treatment for endometriosis, at least 25 percent of patients will have a rapid recurrence after discontinuation of the medication.

Q. I have noticed some deepening and changing of my voice while I was taking Danocrine. Is this a common side effect?
A. Yes, indeed. It has been reported that up to 80 percent of patients taking Danocrine will show some side effects due to its androgenic nature. These changes include increased hair growth, the presence of acne, the presence of oily skin and, as you noticed, deepening of the voice. Unfortunately, some of these could persist even after the patient discontinues the medication.
Another side effect of Danocrine is the changes it can produce in the liver enzymes and the serum level of lipoproteins, the so-called LDL and HDL, which are low density lipoprotein cholesterol and high density lipoprotein cholesterol. Therefore, there is some concern about arteriosclerosis in patients who are taking Danocrine.

Q. Lately, I have been hearing more and more about GnRh agonist and I was told that this may be a promising treatment for endometriosis. Is this true?
A. The growth of endometriosis is dependent upon the level of estrogen in the blood (the major hormone produced by the ovarian follicle). The production of this hormone is stimulated by two hormones from the pituitary gland, namely FSH (follicle stimulating hormone) and LH (luteinizing hormone). FSH and LH are called gonadotropins. The production of these hormones is also controlled by another hormone that is produced in the brain: GnRH, which is the gonadotropin releasing hormone.
It has been shown that the presence of GnRH in blood stimulates the production of LH and FSH. Also, the administration of these hormones exogenously in one pulse (one dose) per hour is associated with the secretion of LH and FSH. However, the administration of GnRH in a rate of more than three pulses per hour, or in a continuous fashion, produces an initial increase in LH and FSH which is followed by a drastic decrease in LH and FSH. Therefore, on the basis of this finding, the continuous administration of GnRH agonists has been used for the treatment of endometriosis in the past several years.
Initial investigation has shown that they are effective in relieving symptoms of endometriosis within a month of administration. Studies which were conducted after diagnostic laparoscopic confirmation of endometriosis, and after treatment, showed that besides the regression of symptoms, the GnRH showed significant reduction in the size of endometrial implants after therapy.
Lupron (luproid acetate) and Synarel (nafarelin acetate) are the two agonists approved by the FDA for treatment of endometriosis. Lupron is administered in monthly injections (3.75 mg intramuscular) and causes profound pseudo-menopausal status in most, but not all, patients. Synarel is administered as an intranasal spray employing a metered nasal spray system (each spray contains 200 micrograms).
The side effects of GnRH are basically that of low estrogen, and the most common complaints have been hot flashes and other symptoms of menopause, including lack of sleep, mild and transient depression and, in some patients, breast tenderness and decreased sex drive.

Q. I have a friend who was treated in France with a medication named Gestrinone. Is this available in the United States, and how does this work on endometriosis?
A. Although the use of this medication is rather new for treatment of endometriosis, it has been known to us since the 1970s as a once-a-week contraceptive. Gestrinone works by increasing the male hormones (androgens) in the body. Increasing the androgens produces a hostile environment for the ovarian follicle to grow. Therefore, patients who have been taking Gestrinone have blocked follicular production and diminished production of estrogen, which is necessary for the growth and development of the endometrium in endometriosis. The second mechanism by which Gestrinone can affect endometriosis is a decrease in the secretion of the central hormones for stimulation of the ovaries, i.e., LH and FSH.
There are several advantages to this medication. First, it can be used two or three times per week rather than daily, and the side effects are rather less than Danocrine, although some patients have shown signs of acne, headaches, spotting and abnormal uterine bleeding, some mild weight gain and increased hair growth. However, this medication apparently does not alter the blood cholesterol or triglyceride levels. Several studies in Europe and other parts of the world have shown that this medication could result in about 90 percent relief in the symptoms of endometriosis about two months or so after the initiation of treatment. The pregnancy rate in these patients has been shown to be similar to that of Danocrine.

Q. My older sister was treated with testosterone for her endometriosis. Isn't this a potent male hormone and are they still using it for the treatment of endometriosis?
A. Yes, testosterone is actually a male hormone. It was first used about 30 years ago for treatment of endometriosis through a direct androgenic effect. Although testosterone supplies effective relief against the symptoms of endometriosis, its fertility rate has not been as good as other treatment modalities, e.g., danazol. However, we have been reluctant to use testosterone for treatment of endometriosis due to its potent androgenic action and its side effects, which are very potent. Another danger of using testosterone is that patients can get pregnant while being treated with it and could inadvertently continue to use it during pregnancy with some significant side effects to the fetus.

Q. What is the role of oral contraceptives in the treatment of endometriosis?
A. Oral contraceptives were one of the first medical treatments for endometriosis. Indeed, the first report for the use of these medications came out about 35 years ago and the scientific basis for that was the production of so-called pseudo-pregnancy due to these hormone combinations. Birth control pills are a combination of estrogen and progesterone, and they function by producing anovulation, or stopping ovulation in the patient. The biggest problem here is that since there is estrogen in these pills, it makes sense that the estrogen may have the reverse effect on endometriosis and actually stimulate activity in the endometriotic lesion. That is why contraceptive agents that have been recommended for the treatment of endometriosis are the ones with low estrogen and high progesterone. Studies which have compared the effects of oral contraceptives with danazol treatment for endometriosis have shown that it is half as effective as danazol in relieving the symptoms, and the major side effects are much like other oral contraceptives (consisting of breakthrough bleeding, nausea and vomiting).

Q. My mother was given a medication named progestin. What is this and how effective is it in the treatment of endometriosis?
A. Progestin is synthetic progesterone and in fact was one of the very early hormonal treatments for endometriosis. In general, it works on FSH and LH by inhibiting their production and secretion, and causing an acyclic low estrogen environment. Obviously, this would affect the growth of the endometrium and endometrial implants. In addition, since all of these agents contain some weak androgenic activity, they also have been shown to have some direct effect on the tissue.
One reason these medications did not really achieve popularity and are not being used at the present time as a first line agent for the treatment of endometriosis is that there are severe side effects, such as bloating, weight gain, depression and irregular bleeding. They also require large doses to be effective, and these large doses occasionally produce a rather prolonged effect on the ovarian production of estrogen. After treatment, some patients have a prolonged history of amenorrhea (not having periods).

Q. What do you think about the no treatment option?
A. In the past, and prior to the use of laser treatment, the expectant therapy or no therapy was an option in some patients with mild or minimal endometriosis. Since up to 60 percent of these patients were getting pregnant within two years of follow-up, this method was an accepted way of follow-up and treatment. However, we believe that since diagnosis of endometriosis is usually made at the time of laparoscopy, lasers should be used to treat any endometriosis that is found and, if pregnancy does not occur in nine to twelve months, other treatment modalities can be applied.

Q. Can endometriosis be decisively diagnosed by pelvic exam?
A. As with symptoms, the clinical signs of endometriosis during a pelvic exam may be nonexistent, minimal, or marked as a function of location and total mass of the disease. However, there are clinical signs that can increase the index of suspicion in patients with symptoms of endometriosis: thickness and feeling of nodularity in the posterior pelvic area; pain and tenderness during pelvic examination; fixation or relative decreased mobility in the tubes or the ovaries due to the presence of pelvic adhesions; presence of a uterus tilted backward and, in more advanced cases, due to the presence of endometrioma, feeling of pelvic mass. However, none of these clinical signs are a decisive sign of endometriosis, and final diagnosis can only be confirmed by laparoscopy.

Q. Who is a candidate for conservative surgical treatment?
A. In the past 20 years, and particularly in the past 10 years, conservative laparoscopic surgery has become the treatment of choice for most cases of endometriosis (more than 90 percent). It has repeatedly been reported that results of laparoscopic surgery are at least as good, if not better, than the results of laparotomy. This is in regard to economic considerations (decreased hospital time), cosmetic considerations (lack of a long scar), and the convenience of a quicker recovery. Because of this, patients have themselves made this method the preferred first line of treatment. We think that very few patients with endometriosis may not be manageable via laparoscope.
Key to the success of this therapy method:
1. Any and all visible lesions that can be removed safely must be removed, including nonpigmented and the so-called atypical endometriosis. 2. Any suspicious lesions, particularly on the ovaries, should be biopsied.
3. In cases where all the lesions cannot be removed, efforts should be made to remove as much disease as possible (so-called debulking) to give a better chance for post-operative medical therapy.
4. Although it has been reported that the success rate (relief of pain and/or pregnancy) following laser surgery is not any greater than other types of treatment (i.e., electrocautery, blunt dissection or sharp dissection), I personally prefer the use of laser therapy because we can safely destroy the endometriotic lesions next to or even on sensitive pelvic structures, such as the bladder, bowel and ureter, without damaging the normal tissue.

Q. How successful is laparoscopic surgery? A. The results obviously depend on the patient's initial complaints and also on the type of the procedure that is performed. The reports available in clinical studies in the past two decades indicate that removal of endometriosis and ablation, along with lysis of adhesion, can give relief of the pain to 70 to 90 percent of the patients, with a 10 to 20 percent rate of recurrence within three years of the treatment.
However, when the patient's main concern has been fertility, studies of more than 1000 patients have found a greater than 65 percent pregnancy rate in mild endometriosis, 50 percent in moderate, and 40 percent in severe cases of the disease. An interesting observation is that approximately 75 percent of patients who conceive after a conservative surgery do so within the first 12 months of the surgery. Since the outcome of the surgery and the pregnancy rate are inversely related to the stage of the endometriosis, early detection and treatment is the key to higher success.
It has also been reported that post-operative medical therapy in all subjects gives a superior pregnancy rate. Also, in our experience (in more than 1000 cases of endometriosis), the CO2 laser has been a superior laser for treatment for pain relief and for restoration of fertility.

Q. Do you have any recommendations for increasing the pregnancy success rate after conservative surgical treatment? A. One should make sure that there are not any other factors besides endometriosis which were contributing to the couple's infertility. Such factors might include cervical factors, sperm disorders, or luteal phase defect, among others. If such factors are present, they should be corrected.
It has also been suggested and reported that post-operative medical therapy could result in a superior pregnancy rate. However, our philosophy has been that since the first nine to twelve months after a complete removal and vaporization of the endometriosis is the most critical and productive time (75 percent of the patients who get pregnant following surgery do so during this time), medical treatment can be initiated after this period. In cases where a complete removal was not possible during the surgical procedure, immediate post-operative medical therapy can result in far superior success rates.

Q. What is the role of assisted reproductive techniques such as in vitro fertilization (IVF) and gamele intra fallopian transfer (GIFT) in the treatment of infertility of patients with endometriosis?
A. In the past 10 years, and more specifically in the past five years, IVF and GIFT have been documented to be of great benefit in patients with long-term infertility and endometriosis that has failed surgical and medical therapy. IVF and GIFT have shown, overall, a 25 and 30 percent pregnancy rate, respectively. In other words, there is no difference in pregnancy rates for endometriotic patients when compared with other groups of patients using these techniques.

Q. How often does endometriosis recur following treatment?
A. In different reports, the recurrence rate of endometriosis following surgical and/or medical therapy is 10 to 20 percent. However, there are a few points which should be kept in mind when considering these figures:
1. Pregnancy following any treatment will decrease the risk of recurrence dramatically. In one study, the rate of reoperation was four percent in patients who conceived after treatment, and 40 percent in those who did not.
2. Although the rate of pain relief is more than 60 percent after a second operation, the pregnancy rate after the second procedure is approximately 25 percent. 3. The more advanced the stage of endometriosis at the time of the initial surgery, the higher the rate of recurrence.
4. Although the mean time of reoperation is about three years after the initial surgery, new studies and "second look" laparoscopies have confirmed the recurrence of active endometriosis after only a few months following therapy in some patients.
5. Not all patients who have recurrence of pain and other symptoms were found to have endometriosis in the "second look" laparoscopy. Interestingly enough, only 30 percent had documented endometriosis, and the majority were found to have pelvic adhesions or other causes of pelvic pain.
6. In infertile patients with recurrence of endometriosis, particularly in patients over 35 with a long history of infertility and more advanced stages of endometriosis, IVF could be a better substitute to a second surgery if the initial surgery or medical treatment fails.

References

Babaknia, A. Laparascopic CO₂ laser treatment of infertile patients with endometriosis who failed to respond to initial medical or surgical therapy. Proceedings of the 38th Annual Meeting of the Pacific Coast Fertility Society, p. 27, 1988.
Corson, S. L., M.D. Endometriosis: The Enigmatic Disease. Canada: Essential Medical Information System, 1992.
Is mild endometriosis a disease? (Debate). Human Reproduction, 9:2202-2211, 1994.
Martin, D. C., et al. Laparascopic appearance of peritoneal endometriosis. Fertil. Steril., 51:63-67, 1989.
O'Connor, D. T. Endometriosis. New York : Churchill Livingstone, 1987.
Shaw, R. W., ed. Endometriosis. Park Ridge, N.J.: Parthenon Pub. Group, 1990.
Wilson, E. A. Endometriosis. New York : Liss, 1987.