Hyper Pro Lac Tine Mia

Hyperprolactinemia David Stanford, M.D.

1

I.

Neuroendocrine Regulation of Prolactin (PRL) Secretion A.

Dual hypothalamic regulation 1.

PRFs: TRH, VIP, PHM

2.

PIFs: dopamine is primary–possible role for GAP (GnRHstriated peptide)

3.

PIF activity is dominant; PRL is under tonic inhibition by hypothalamus. If the stalk is cut, PRL levels rise whereas other hormone levels fall.

B.

Primary target organ is the breast: suckling stimulates afferent pathways through cord to elicit PRL release in puerperium

C. Metabolic factors: arginine and hypoglycemia stimulate D. Estrogen stimulates lactotrophs directly E. II.

PRL is secreted episodically with nocturnal surge

Manifestations of Hyperprolactinemia A.

Galactorrhea indicates elevated PRL in 10% of women and 99% of men

B.

Amenorrhea: indicates elevated PRL in 15% of women

C. Galactorrhea plus amenorrhea: indicates elevated PRL in 75% of women D. Infertility: indicates elevated PRL in up to 33% of women E.

Osteoporosis: increased with elevated PRL--due to estrogen lack. If normal menses are present, osteoporosis does not occur.

III.

Mechanisms of Reproductive Dysfunction in Hyperprolactinemia A.

Inhibition of pulsatile GnRH secretion

B.

Interference with gonadotropin action in ovary

C. Interference with estrogen positive feedback D. Inhibition of FSH-directed ovarian aromatase E.

Inhibition of progesterone synthesis by granulosa cells

F.

Inhibition of 5-alpha-reductase enzyme in men, thereby decreasing the conversion of testosterone to DHT

IV.

Differential Diagnosis of Hyperprolactinemia A.

Medications: neuroleptics, methyldopa, metoclopramide, MAO

inhibitors, tricyclic antidepressants, verapamil B.

Pregnancy

C. Hypothyroidism D. Renal insufficiency E.

Cirrhosis

F.

Neurogenic: breast, chest wall, spinal cord lesions

G. Hypothalamic disease: tumors, sarcoidosis, non-secreting pituitary tumors, neuraxis irradiation, stalk section H. Empty sella syndrome I.

Acromegaly

Prolactin is secreted by the pituitary and there is a regulatory mechanism by the hypothalamus so there are prolactin releasing factors like basal active intestinal peptide, polytropin releasing hormone, PHM which is a prepart of the precursor to VIP. All of these are prolactin releasing factors. The primary prolactin inhibitory factor is the neurotransmitter dopamine. There is also a substance a called GAP, which is gonadotropin releasing hormone associated peptide, which is part of the precursor to GnRH. It has been shown to have some inhibitory properties to prolactin. Dopamine is inhibitory. This dual regulatory mechanism within that the prolactin inhibitory component is dominant so that if you cut the hypothalamic pituitary stalk and give it influence of the hypothalamus prolactin levels rise because of this inhibition where as all other pituitary hormone levels will fall. You have the paradox of decrease in pituitary function along with hyperprolactinemia. Prolactin has as its primary target organ the breast to facilitate milk production as a puerperium. There are a number of prolactin receptors on other tissues, but we don’t know their precise physiologic functions. A stimulus by the neonate will cause the stimulation of afferent pathways going back up to the hypothalamus to accentuate prolactin secretion and within the hypothalamus serotonin is positive, dopamine is inhibitory and medication that lurk within the hypothalamus to decrease dopamine such as the neuroleptic agents and tranquilizers can cause an increase in prolactin for the disruption of the dopaminergic pathways. Prolactin secreted episodically so you see a rise and fall in a single individual over the course of the day with a nocturnal rise. In this single individual that was sampled these levels are superimposed upon the range for five normal individuals. Some of these normal individuals can have a spike of prolactin up to 25-35 mcg per ml. For most laboratories the upper limit of normal is about 25 or 20 so that you can have an occasional prolactin spike even in a normal individual to above normal levels and what you really need to do is to measure 2 or 3 levels in that kind of person to be sure that they do indeed have sustained hyperprolactinemia when this single screen comes back at a borderline level. Manifestations of hyperprolactinemia: There is an alteration of reproductive function so those women may present when they are hyperprolactinemic with amenorrhea or galactorrhea. If we flip this around and look at a patient who presents with a chief complaint of amenorrhea alone about 15% will be found to be hyperprolactinemic. Of those women who come in with a chief complaint of galactorrhea about 25% or so will be found to be hyperprolactinemic, although this doesn’t really include the patient that you are doing a routine physical exam and find some breast fluid. That doesn’t count on that list if you look at all women who are shown to have galactorrhea on examination only about 5-10% will be found to be hyperprolactinemic. Those who have a combination of galactorrhea and amenorrhea ¾ will be found to be hyperprolactinemic. In some series of women who present with infertility up to a 1/3 may be hyperprolactinemic, although clearly this is a mixed nag of patients because many of these women will also be oligo or amenorrhea and will have other manifestations as well. As a result of hyperestrogenemia associated with hyperprolactinemia osteoporosis may occur. This is from the first study that showed that looking at an older technique by looking at bone density in the radial shaft and finding that the levels are quote decreased in women who are hypoestrogenemic. If you have a woman that comes in complaining of galactorrhea who has normal menses and normal estrogen levels then that woman is not at particular risk for osteoporosis and perhaps doesn’t need any treatment at all. Hirsutism may be a problem. This is the famous painting of woman suckling her infant and hirsutism occurs in about a 1/3 of patients with

2

V.

J.

Prolactinomas

K.

Idiopathic

Diagnostic Evaluation A.

Basal PRL levels at least twice: 1.

PRL >200 ng/mL = prolactinoma or renal failure

2.

PRL <200 rig/mi. = prolactinoma or any of the other causes

B.

Routine history and physical, SMA 20 and TSH excludes almost all above except hypothalamic and

pituitary

disease C. CT or MRI to differentiate hypothalamic/pituitary disease from idiopathic, even with (anything > 25 mg/m!.) VI.

Pathogenesis of Prolactinomas A.

Most evidence weighs against a hypothalamic etiology 1.

No hyperplasia in areas of pituitary distant from tumors

2.

Secretory dynamics return to normal following curative selective adenoma resection

B.

3.

Postoperative recurrence rate low and takes place early

4.

Tumors axe monoclonal

Oral contraceptive use now shown to not be associated with conversion of clinically silent to significant prolactinomas

C. Only

5-6%

of

microadenomas

enlarge

to

form

macroadenomas VII. Treatment A.

Idiopathic hyperprolactinemia: bromocriptine is effective in 85%

B.

Microprolactinomas 1.

Transsphenoidal surgery: initial cure rate 80-85%, with a recurrence rate of 20%. Depends on skill of surgeon

2.

Radiotherapy: ineffective and takes a long time

3.

Bromocriptine: restores PRL to normal in 80-85%

4.

Observation only; follow PRL. Repeat CT/MRI if PRL levels rise

C. Macroprolactinomas 1.

Surgery: cure rates <50% and very much dependent on size with recurrence rates 20-50%

2.

Bromocriptine: size reduction to <50% of original size in 50%, to 50% in 16% and to 10-30% in 33% a.

First evidence of size reduction may occur after 6 weeks

b.

Size reduction does not correlate with basal or nadir PRL or percentage reduction in PRL levels

c.

In first 2-3 years, most will reexpand

d.

After a few years, few reexpand

VIII. Pregnancy and Prolactinomas

hyperprolactinemia. PCO and hyperprolactinemia are commonly associated. The precise link as to what is causing what is still not firmly established, but it may be the hyperestrogen levels that are occurring in PCO. Increased adrenal androgens are also produced and may cause the hirsutism. Of course this could have been a man with galactorrhea, and the findings of galactorrhea in a male is pathogenic of a prolactinoma. There are a number of potential sites of action of hyperprolactinemia causing inhibition of reproductive function. The most important of which is here in the hypothalamus where there is a suppression of the pulsatile release of gonadotropin releasing hormone which then causes a decrease in the pulsatile release of FSH and LH. There is also an interference of gonadotropin action. There is an inhibition of positive estrogen feedback at mid-cycle. There is inhibition of aromatase activity within the ovary, and in the male there is an inhibition of 5-alpha-reductase, which is the enzyme that is necessary to convert testosterone to DHT. In an individual who is hyperprolactinemic you see a decrease in pulsations of LH. The patient here had a prolactinoma resected selectively you see the increase in pulse amplitude is shown here. This again is reflecting the increase now in the pulse of GnRH that is occurring. One thing that you should be aware of is that these tumors not only alter reproductive function, but they may have other manifestations as well, particularly in a patient who has a large prolactinoma. Any patients that have larger tumors you have to be aware that these patients may have a variety of mass effects due to the pituitary tumor, so they may present with headaches and visual field defects. They may have varying degrees of hyperpituitarism that need to be evaluated and treated. They may have diabetes insipidus and other types of hypothalamic dysfunction depending upon the size of the tumor. They may have impingement of cranial nerves within the cavernous sinus. A variety of other things can occur due to effects of the tumors, and these must be addressed as well in evaluating patients who have larger tumors. Let’s look at the differential diagnosis of patients who are hyperprolactinemic. I think we have all been fooled by patients who are amenorrhea and hyperprolactinemic and are actually pregnant at the time so that this is probably the most common cause of hyperprolactinemia and amenorrhea, and you are aware of the rise of prolactin that occurs throughout gestation reaching levels of over 200 mg per ml in many women at the time of delivery preparing the breast for milk production. Some women who breast feed don’t change from the usual pattern of a demanding breast feeding schedule to a scheduled type of breast feeding schedule in the first months or two. When they do make that normal transfer, prolactin levels rapidly fall after delivery over the first couple of months and then are no longer stimulated by the suckling stimulus. On the other hand women who had 4 months postpartum still has 9 suckling episodes through the day, continues to have a rise in prolactin with each episode and essentially has to stay in hypoprolactinemia. In many women who continue to demand breast feeding for even months and years postpartum they have continued hyperprolactinemia and continue the impairment of gonadotropin secretion and they remain amenorrheic. A number of developing countries still use this as a method of birth control, albeit it is certainly not 100% successful. Again you need to look at the breast feeding history in a postpartum women that is bound to be hyperprolactinemic. This is the last new patient that I saw with hyperprolactinemia referred from an obstetrician who was seeing this patient who was still having hyperprolactinemia with persistently despite bromocriptine treatment and in fact when I went in to see it was clear that this lady had acromegaly. You need to be aware of patients who have acromegaly may secrete prolactin in about 40% of cases. Renal insufficiency can also cause hyperprolactinemia. These are

3

A.

No teratogenicity or other untoward effects on fetus of bromocriptine in >6,000 pregnancies

B.

Risk of symptomatic microadenoma enlargement: 1.6%

C. Risk of symptomatic macroadenoma enlargement: 15.5% if no previous surgery/irradiation but only 4.3 % if previous surgery/irradiation. Options: 1.

Stop bromocriptine when pregnancy diagnosed and

patients on hemodialysis and on CAPD. You can see the quite substantial prolactin elevations that some of these patients have. Lesser degrees of renal insufficiency can also do this so that in women who have creatinine levels between 2 and 10, many of these woman have elevated prolactin levels, and they are also taking medications that are known to alter central neurotransmitter regulation so you can get an exaggerated prolactin rise. Women who are using metoclopramide for control of vomiting, thiazines, Aldomet, and any hypertensives like verapamil these drugs can raise prolactin and in the setting of renal insufficiency can cause a dramatic increase.

observe. If tumor enlarges, reinstitute bromocriptine----if fails, surgery

IX.

2.

Operate on tumor prepregnancy to allow room to enlarge

3.

Continue bromocriptine throughout pregnancy

Special Considerations A.

Tumor fibrosis: primarily a problem for macroadenomas in that it may decrease later surgical cure rate. If tumor shrinks bromocriptine should be continued.

B.

Long-term bromocriptine: taper and try to discontinue

C. Growth of tumor while on bromocriptine: noncompliance or

In a series of patients from England where they looked at a number of patients who had mild to moderate prolactin elevation, six of these women had macroadenomas, and these were non-secreting tumors causing prolactin elevations. I am always asked the question how high does the prolactin have to be to do a CT or MRI scan, and the answer is it just has to abnormal at all. Prolactin of 25 or 40 all of these demands an MRI scan. Once you have excluded the readily diagnosable causes that we find on history and physical. There are a variety of medications that can interfere with dopamine in the hypothalamus and can cause prolactin elevations. She can have chronic chest wall irritation from a burn or a malignancy. Patients who have hepatic cirrhosis about 10% will get hyperprolactinemic. Renal insufficiency is quite uncommon, but those women can be hyperprolactinemic as well.

possible carcinoma or hemorrhage into tumor D. Options for patients still hyperprolactinemic after surgery who do not respond to bromocriptine 1.

Other dopamine agonists: cabergoline (Dostinex) is well tolerated, once weekly dosing, pergolide (Permax), is once daily

E.

F.

2.

Reoperation

3.

Irradiation

Intolerance to bromocriptine 1.

Try intravaginal bromocriptine: no nausea and vomiting

2.

Try cabergoline

We are left with lesions that are within the hypothalamus and pituitary, which demand an MRI scan or CT scan to evaluate these mass lesions. Once you have excluded these other causes you really must go to CT or MRI to exclude these causes. They are relatively common in the causes of hyperprolactinemia. With an MRI scan you get a very clear view of the hypothalamic pituitary stalk as well as the chiasm. One of the things that we have to consider is that in fact if you do that type of procedure in normal individuals you find tumors quite commonly. In fact if you do autopsies you find that macroadenomas can be found in 11% of so-called normal individuals. If you look around this room 10-11% of you have macroadenomas in your pituitary. The cause could be from medications so that you can be fooled. You just have to be aware that this does occur.

Concomitant estrogen use: safe for almost all patients. Must follow PRL levels to detect the rare patient that may have an estrogen-induced increase in tumor size

There really has been two major hypotheses as far as the pathogenesis of hyperprolactinemia and prolactinomas. The tumor arises because of underlying hyperthalemic dysfunction perhaps to much releasing factor or too little dopamine so that initially you get some hyperplasia going on in the pituitary with eventual transformation to a tumor. The alternative hypothesis is that the tumor arises with little regard to what is going on here in the hypothalamus. There are really a number of ways of addressing this particular questions. You can look at the pituitary tumor away from the area of the tumor to see indeed if there are areas of hyperplasia. In fact there are not areas of hyperplasia suggesting that this tumor essentially starts in the pituitary. The second thing that you can do is to see what happens if you take out the tumor. Most patients who have their tumor successfully removed remain with a normal prolactin level. Indeed some patients have a recurrence of hyperprolactinemia and that recurrence occurs usually within the first year or so following surgery. To me this means that there is probably some re-growth of tumor remnants left behind by the neurosurgeon rather than new tumor formation because we know that these tumors grow very slowly. The third line of evidence is looking at the abnormal prolactin secretory dynamics that occur in patients who have prolactinomas such that the prolactin response to TRH. Patients preoperatively who have blunted prolactin responses to TRH if they are still not cured by surgery they still have a blunted response. If they are cured by surgery they now have a normal doubling or tripling of prolactin levels. I am not recommending that you TRH tests anymore but in days when this was done we tried to sort out what the path of physiology was. Finally we have studies from molecular biology that look at the clonality of the tumors. If you recall the Lyon hypothesis from embryology suggests that in certain tissues in cells either maternal or paternal x chromosome gets oppressed so that you have some cells reflecting maternal chromosome and some reflecting the paternal x chromosome. Most tissues have a combination of both of these cells. You have genetic mosaics in the adult tissue. If this tissue

4

becomes hyperplastic and then forms a tumor what you find is a polyclonal tumor having both maternal and paternal x chromosomes. Alternatively if there is a single cell that mutates and then multiplies you would then find a tumor that is monoclonal either expressing maternal or paternal chromosome. In fact when you look at pituitary tumors for their clonality you find that they are indeed monoclonal tumors rather than polyclonal tumors for virtually all that has been evaluated suggesting that in fact this is a single cell that mutates within the pituitary causing that tumor to proliferates. There has been a intense hunt to look for the pathogenic mutation within the cells. Sites have been looked at for potential sources of the mutation. None of these at the present time have really been found to be a particular mutation at increased frequency unlike the situation with acromegaly. We know that 10-11% of people have pituitary adenomas in this room. One other thing to realize is that these tumors grow very slowly. In fact most microadenomas that we see in our patients don’t grow at all. They kind of sit there without any evidence of growth. The progression rate was only 6.9% of evidence of tumor growth as judged by CT scan primarily in these particular studies. The follow-up here is anywhere from 2 to 8 years following these patients. Most patients that have microadenomas stay with that. Clearly the microadenomas had to come from that, but it is a small proportion that make that transition. Clearly these are the goals that we wish for any type of tumor. We would like to have a hormonal cure. We would like to reduce the mass and perhaps eliminate the mass or the tumor itself. Any visual field defects or other sequela of the large mass we want to correct to prevent any recurrence and preserve normal pituitary function. Transsphenoidal surgery is certainly one mode of treatment. The skill of the neurosurgeon in doing transsphenoidal surgery is really critical in nature. In the best of neurosurgical hands what you can expect from transsphenoidal surgery a normalization rate of prolactin levels in about 70-75% of your patients. The recurrence rate is anywhere from 15 up to 50%. If you subtract this from this the normalization rate is about 50-55%. The mortality rate is about a ¼ of 1%. The morbidity rate is just under a ½%, although these levels are quite low they are not zero. For patients who have adenomas and the tumor is greater than a centimeter in diameter the initial normalization rate of prolactin is 32% and recurrence rate is 19% so that the long-term cure rate is 13%. Even when you look at these numbers you realize that patients that have very large tumors often don’t get operated on because of the nature of that type of surgery. The mortality rate is here is just under 1% with a much greater increase in morbidity rate. Again these are numbers that you have to use when discussing therapeutic options with your patients. Radiotherapy can also be used as primary therapy. What I have shown here are prolactin levels where radiotherapy was done as primary therapy. The prolactin levels that I have given here in milliunits per liter so that in fact out of these patients you can see about 1/3 of them getting normal prolactin levels, but it takes between 3 and 12 years to get there. If you have a 38-year-old woman who wants to get pregnant tomorrow radiotherapy is certainly not going to be a useful modality for that patient. This is the reason why we really reserve radiotherapy for patients who have not responded to either surgical or medical treatment. There are a lot of adverse effects of radiotherapy. One is hyperpituitarism which occurs in at least 50% of patients as you follow them out over the years. The second is brain tumors which occur in about 2% of patients like 20 years post radiotherapy. These are often malignant tumors. You may get cranial nerve damage. Strokes are increased in frequency. There may be hemorrhaging to tumors. Finally cognitive dysfunction occurs in some patients. There is not good data on how this frequently this occurs but for those of us who have seen a lot of patients and had radiation it seems to be a rather frequent phenomenon occurring anywhere from 10-15% of patients. Let’s turn now to medical treatment, which is what you are most familiar with I think and that is using bromocriptine. Bromocriptine is a long-acting dopamine agonist working at the dopamine receptors in the pituitary to shut off prolactin secretion. If you look at the data on using bromocriptine we find that you can get a normalization of prolactin in about 80-90% of patients. There is about 5-10% of patients who do not respond to it. There is another 5-10% of patients who are intolerant to it because of adverse side effects. We also can by shutting down the protein synthesizing machinery within the cells we can cause those cells to involute and cause a tumor size reduction for the larger tumors. You get a 50% reduction of tumor size in about 40% of patients. You get a 25-50% reduction in tumor size in about 28% of patients. There is less than a 25% reduction in 11% and no change in tumor size in about 18% of patients. Again these are numbers that you can use in discussing therapeutic options with your patients. There clearly are adverse effects in using bromocriptine. Many times they can be prevented by starting very slowing and building up over a period of a few weeks. There are many patients that respond to 1.25 mg b.i.d., so that we really start checking prolactin levels on that low dose because patients do very nicely sometimes with that. Nausea and vomiting is common and sometimes patients just don’t tolerate it. Orthostatic hypotension can be a problem and again as you see here there are a variety of other less common side effects that for some patients may be limiting. Because of this we have to figure out other ways of getting around these side effects if we are going to treat the patient. One way is to give the oral bromocriptine tablet intervaginally. By doing so you can get a slow absorption of the bromocriptine, but in fact you do absorb it, and you will get a very nice lowering of prolactin levels. For the patient who is having GI intolerance you can often give the tablet intravaginally with very nice reduction of prolactin levels. It only has to be given once a day this way rather than two or three times a day because of the slow absorption rate, and it really can be very useful for some patients who don’t tolerate the medication because of GI side effects. Nausea and vomiting is a local GI effect for most patients. Pergolide is available. The brand name of this is Permax, to use to treat patients with hyperprolactinemia. There clearly are some patients that respond to pergolide who do not respond to bromocriptine and vice-versa. The major problem with pergolide is that we do not have clinical experience in using it in women who wish to become pregnant, so there is a safety data for use during pregnancy. Cabergoline or Dostinex is now available in the pharmacy. This drug appears to be more successful in use than bromocriptine so that you have a greater proportion of women in this randomized study who had ovulatory cycles. There was a lower amount of adverse effects, and a small amount of women who discontinued the drug for adverse effects. It also can be given once a week rather than several times a day. With once a week therapy you can get excellent lowering of prolactin levels with virtually no side effects. The package says that it should be given twice a week. You can do fine with once a week. The patients will do very well. If the patient does have side effects by splitting the dose of twice a week you can sometimes get rid of that. I really have not seen a patient not be able to tolerate the drug. It also will cause tumor size reduction. In this study that we did many of these patients had already received bromocriptine or other dopamine agonists previously and had been found to have adverse effects with those drugs. It is a very well tolerated drug. It works perhaps a little better than bromocriptine with less side effects. If we compare the overall success rate in decreasing tumor size of these drugs there is a comparable decrease in tumor size between all of the drugs. There are still other alternatives besides dopamine agonists and that is the use of oral contraceptives.

5

We are conditioned to the past to think that elevated estrogen levels from pregnancy may cause tumor stimulation. This has sort of limited us in the past from giving estrogen replacement to women because we were afraid it might stimulate tumor growth. It is a very safe type of thing to do. There are clearly going to be some patients that are going to have an estrogen induced prolactin increase in these patients. If you are following a patient you really have to follow prolactin levels if you put them on estrogens. I have seen a couple such women with a prolactin level that went from about 150 up to 400 or 500. I stopped the medication and got an MRI scan and found no change in tumor size, but I was not going to try risking continued therapy in that type of a patient. You just have to be careful about that kind of a patient, but I think estrogens are quite safe to use as long as they are done carefully. We also should remember that we don’t absolutely have to get prolactin levels down to achieve fertility. There are other ways of doing this. There is a variety of methods of ovulation induction that can be done. You can treat men as well. Well let’s complete this now and look at pregnancy. Bromocriptine when used to facilitate ovulation obviously what you are going to do is to give it and get the prolactin levels down to normal. At the time that she has missed her period you do a pregnancy test and that is the time you stop the bromocriptine. Therefore, the fetus has been exposed to bromocriptine under those circumstances probably for three, four or five weeks. When bromocriptine is used in this fashion we now have data on over 6,000 pregnancies, and there is certainly no problems as far as spontaneous abortions or atopic deliveries, pre-term deliveries, multiple births, or malformations or perinatal disorders. these kids who have been born of women who are taking bromocriptine have been followed up to 6 or 7 years of age. There development is completely normal. They don’t have anything wrong with them, so from a child’s point of view it is a perfectly safe drug to use. There is no increase in spontaneous abortions, tubal pregnancy, pre-term deliveries, and no increase in multiple births when using cabergoline. There has been a total of 5 major malformations out of 208 pregnancies so that it is just under 2½ percent, so this is within the expected range. However, it is a small number of patients that have been delivered. It is probably safe, but we don’t have large numbers. We don’t have a lot of data that I would feel very comfortable in telling a women if she is taking cabergoline to just go ahead and get pregnant. I would rather switch her to bromocriptine or at least outline the risk that we just don’t know for sure. On the other hand this data I think is reassuring that if a patient does get pregnant while on cabergoline there is certainly no reason to go ahead with a therapeutic abortion. There is a question arising for the mother. There is a patient of mine who had a macroadenoma that was just at the borderline about 1.1 cm she stopped her bromocriptine and became pregnant and at 7 months of pregnancy was complaining of increases in headaches. An MRI scan performed at this point shows an increase in size of this macroadenoma. Pregnancy can clearly cause an increase in size of tumors. If we looked at patients who got microadenomas, the risk of tumor enlargement here is just over 1%. This is not a major problem. When we are talking about enlargement here we are talking about clinically significant enlargement. On the other hand, patients who have macroadenomas the risk of significant tumor enlargement is about 15%. If you do some surgery or radiotherapy to this women prior to pregnancy you reduce this considerably but don’t eliminate it entirely. These are the patients that are in greatest risk, and they are the ones that we need to address our attention to. Finally in following patients who are hyperprolactinemic with microadenomas there really is not much change over the course of gestation. It doesn’t help very much to check her prolactin through pregnancy in this type of a patient. In addition you can have a patient whose tumor enlarges without much in serum prolactin levels so that these patients must be followed clinically. In patients therefore, who have microadenomas I think what we really need to do is to establish an intermenstrual interval, stop bromocriptine when the menses have been missed and then pregnancy test is positive. Patients are followed symptomatically. If there are any types of headaches or visual complaints, then we proceed with an MRI scan and visual fields. If the tumor does enlarge significantly then you can re-institute bromocriptine to cause the tumor to shrink at that point in time, and it virtually always works. The much more problematic patients who has a big macroadenoma who often had a lot of extension, and the question is what do you do about this patient. If the patient has had considerable tumor shrinkage with the bromocriptine, then that is really a particularly problematic issue. One thing that you can do is to do surgery prior to pregnancy to de-bulk the tumor. Another approach is to stop the bromocriptine and then watch the patient very carefully and see them on a monthly basis checking visual fields. Certainly if there is any suggestion that the tumor is enlarging then you do an MRI scan, then reinstitute bromocriptine if the tumor enlarges and clearly if the patient doesn’t respond with a decrease in size of the tumor to bromocriptine then surgery would be indicated to de-bulk that tumor. A third approach is to continue bromocriptine throughout the pregnancy if they have had substantial tumor size reduction and again what we have here are relatively limited data. It is just over 100 patients that have been treated throughout the pregnancy and that data shows that there is no increase in malformations and adverse effects but again the data is relatively sparse.

6