British Journal of Obstetrics I I I

I

and Gynaecology

VOL 84 No. 4

NEW SERIES

APRIL 1977

MANAGEMENT OF HYPERPROLACTINAEMIC AMENORRHOEA BY

S. FRANKS The Middlesex Hospital, London H. S . JACOBS

M. G. R. HULL* St Mary’s Hospital Medical School, London S. J. STEELE AND

J. D. N. NABARRO The Middlesex Hospital, London Summary Results of treatment of 52 patients with amenorrhoea associated with hyperprolactinaemia are presented. All patients had a detailed radiological examination of the pituitary fossa, including lateral tomography in every patient and air encephalography in those in whom a pituitary tumour was suspected. There were 17 patients with untreated pituitary tumours, 5 patients with previously treated pituitary tumours and persisting hyperprolactinaemia, and 30 patients with normal pituitary radiology. Patients with pituitary tumours were treated either by transsphenoidal or transfrontal surgical extirpation of the tumour, followed, if necessary, by external irradiation and/or bromocriptine. Four patients were treated with external irradiation as primary therapy, and three patients who did not wish to conceive were treated with bromocriptine as primary therapy. Patients with normal radiological appearances were treated with bromocriptine as primary treatment. Ovulatory menstrual cycles developed in 42 patients and there were 19 pregnancies. Those ovulating but not conceiving had adequate nonendocrine factors to account for the disparity. Failure of response was seen in 10 patients and was due to inadequate fall of prolactin in response to surgery (2 patients), external irradiation (3 patients) and bromocriptine (1 patient), and gonadotrophin deficiency which developed after surgery in 3 patients but was present pre-operatively in 1. The relative merits of treatment by surgery, external irradiation and bromocriptine are discussed and a policy of treatment outlined.

HYPERSECRETION of prolactin is an important cause of anovulatory infertility. It occurs in 15 to 20 per cent of patients with amenorrhoea, although only about a third of these have galactorrhoea (Franks et al, 1975). There are

*

several causes of hyperprolactinaemia but the most important is a prolactin secreting pituitary tumour (prolactinoma) which occurs in at least one-third of hyperprolactinaemic patients who have been referred directly to a gynaecological endocrine clinic because of amenorrhoea (Jacobs et a/. 1975).

Present address: Bristol Maternity Hospital, Bristol. 24 1 10

242

FRANKS, JACOBS, HULL, STEELE AND NABARRO

TABLE I Treatment of patients with hyperproIactinanaemicamenorrhoea Patient group

Total number investigated

1. Normal X-rays of pituitary region 2. Pituitary tumours 3. Pituitary tumours studied after previous ablative therapy

45

Totals

90

External irradiation

Bromocriptine

Surgery Primary

Secondary

Primary

Secondary

0 10

0

36

4

0 2

29 3

0 3

9

(7*)

(2*)

0

0

5

4

2

32

8

10

* Previous treatment; results reported by Jacobs et a1 (1976) The results of 52 patients (including 5 in Group I1 who have had primary and secondary treatment, and one Patient who received Thyroxine) are described in this paper. The characteristic reproductive disturbances associated with hyperprolactinaemia, including primary or secondary amenorrhoea, symptoms and signs of oestrogen deficiency and failure to ovulate in response to clomiphene, are readily reversed when serum prolactin concentrations are lowered to within the normal range (Franks and Jacobs, 1976).There are now several effective remedies for patients with this condition but the choice depends upon the underlying cause. In this paper we review the management of 52 patients out of a total of 90 we have investigated with hyperprolactinaemic amenorrhoea (Table I) who have so far completed at least their primary therapy. The relative merits of surgical treatment, external irradiation and treatment with bromocriptine are discussed and our current policy of management is outlined.

PATIENTS The 52 treated patients were divided into 3 groups (Table I). Group 1 consisted of 30 patients in whom skull X-ray, coned and lateral tomographic views of the pituitary fossa were all normal; none was taking any drug known to raise prolactin levels and all but one were clinically and biochemically euthyroid. With the exception of this patient, who was given replacement treatment with thyroxine, they were all treated with bromocriptine as primary therapy. Group 2 consisted of 17 patients with pituitary tumours whose primary treatment was surgical in 10, external irradiation in 4 and bromocriptine in 3. Fifteen

of these patients were thought to have prolactin secreting pituitary tumours (prolactinomas), and two non-functioning tumours, which, through suprasellar extension, had caused impaired production (or transport) of prolactin-inhibitingfactor and so hyperprolactinaemia. Group 3 consisted of 5 patients studied only after previous primary ablative treatment for a pituitary tumour, who all received bromocriptine as secondary therapy. The results of some of these patients have been reported elsewhere and the references are given in the tables and text where appropriate. METHODS Hormone measurements Serum prolactin concentrations were measured by radioimmunoassay using reagents kindly supplied by the National Pituitary Agency of the USA. The results were expressed in terms of the VLS 1 preparation. The between-assay coefficient of variation of prolactin measurement was 7 per cent. In normal women studied in our laboratory, the mean (fSEM) prolactin concentration was 7.2 (*0.5) pg/l (n = 50). The methods of measurement and normal ranges of other hormone concentrations reported here have been described (Franks et al, 1975). Dynamic tests Progestogen withdrawal tests (Hull et al. 1975) were performed by administering oral medroxy progesterone acetate (5 mg per day) for 5 days and the results classified as negative (no vaginal bleeding in the succeeding week),

MANAGEMENT OF HYPERPROLACTINAEMIC AMENORRHOEA

positive (normal menstrual bleeding), or impaired (scanty bleeding lasting for no more than 2 days in the succeeding week). The ovulatory response to clomiphene was determined after administering oral clomiphene citrate (100 mg/day) for 5 days (Hull et a/, 1976). The results were classified as positive (ovulation proven by pregnancy or inferred from an increase of serum progesterone concentration to above 25 nM/I, 17 days after treatment was initiated), negative (no change in basal body temperature, no withdrawal bleeding and serum progesterone below 3 nM/l), or impaired (a change in basal body temperature or withdrawal bleeding, but a serum progesterone concentration below 25 nM/l, 17 days after initiating treatment). The acute response to bromocriptine was determined by measuring serum prolactin concentrations in blood samples obtained through an indwelling venous cannula every 30 minutes for 24 hours. All the samples in the acute studies were measured in duplicate in a single radioimmunoassay (within assay coefficient of variation was 4.8 per cent). Long term treatment with bromocriptine was started with a dose of 1 -25 to 2.5 mg of bromocriptine at bedtime and the dose was increased every 3 to 5 days to a maximum of 2.5 mg 3 times a day after meals. Since only one-third of patients with hyperprolactinaemic amenorrhoea have galactorrhoea (Franks et a,, 1975), the index of response to bromocriptine that we have used was the development of ovulatory menstrual cycles

243

during treatment. The patients recorded their basal body temperature and one week after the temperature shift a blood sample was obtained for measurement of serum prolactin and progesterone concentrations. In those not responding (no shift of the temperature, or serum progesterone concentration below 25 nM/l), the prolactin concentration was measured in the same sample and if it was not suppressed to within the normal range (below 15 pg/l) the dose of bromocriptine was raised. Patients trying to conceive were advised to discontinue treatment after the basal body temperature had been raised for 16 days.

RESULTS Clinicalfeatures The clinical features of the 52 patients are summarized in Table 11. Galactorrhoea was present before treatment in 19 patients (36 per cent). Progestogen withdrawal tests were negative in 26 of 29 tested patients (89 per cent), impaired in 2 and positive in 1. Clomiphene tests were negadve in 22, impaired in 6 and positive in 1. Failure to ovulate normally in response to at least one course of clomiphene thus occurred in 97 per cent of the patients tested. The one patient (J.Cr.) who did respond normally (described by Jacobs et al, 1976) had the lowest serum prolactin concentration in this series. She had become pregnant after her first course of clomiphene but was delivered of premature twins who died at 29 weeks of gestation. Three months later she had failed to menstruate. Her serum prolactin concentrations at this time

TABLE I1 Clinical features of the patients given primary therapy

Response to progestogen

Patient group

Total treated

Galactorrhoea

1

30

15

23 negative 2 impaired

2

17

4

3

5

0

2 negative 1 positive 1 negative

Totals

52

19 (36 %)

26 negative (89%) 2 impaired (7 %) 1 positive (3%)

Response to clomiphene 17 negative 6 impaired 1 positive 4 negative 1 negative

22 negative (76%) 6 impaired (21 %) 1 positive (3%)

244

FRANKS, JACOBS, HULL, STEELE AND NABARRO

are shown in Figure 2. During subsequent treatment with bromocriptine she developed regular ovulatory menstrual cycles. One patient had primary hypothyroidism (thyroxine 34 nM/1, free thyroxine index 15, basal TSH over 50 milliunits/l) associated with hyperprolactinaemia (serum prolactin 30 pg/l) and an exaggerated response of both thyrotrophin and prolactin to an intravenous injection of 200 pg of thyrotrophin releasing hormone.

Serum prolactin levels The results of basal serum prolactin measurements before treatment (in all cases the mean of at least 2 measurements) are shown in Figure 1. Prolactin levels were usually highest V(l600) 0

l I

0

-

!

0

W

cn

11

I

GROUP 1

I

GROUP 2

1

I

GROUP 3

FIG.1 Basal serum prolactin concentrations (shown on a logarithmic scale) in 52 patients. The horizontal line indicates the upper level of the normal range in our laboratory. Prolactin levels were usually highest in the patients with untreated pituitary tumours (group 2). Two patients in this group with the lowest levels were thought to have hyperprolactinaemia secondary to (radiologically demonstrated) suprasellar extension of a non-functioning pituitary tumour.

in those with radiological evidence of a pituitary tumour but 2 patients (J.K., and J.McK., Table 111) with large tumours and suprasellar extension had only minor elevations of their serum prolactin concentrations. Results of prolactin measurements before and after surgical treatment in the patients in Group 2 are given in Table 111. RESULTS OF TREATMENT Surgery The results of surgical treatment of 10 patients with pituitary tumours are shown in Table 111. In 6 patients transsphenoidal surgery resulted in a fall of prolactin concentrations to within or close to the normal range. Five resumed ovulatory menstrual cycles without further treatment but the sixth (L.S.), who had had minimal suprasellar extension of the tumour, developed panhypopituitarism after the operation and has not menstruated. Transsphenoidal surgery failed to lower prolactin concentrations into the normal range in one other patient with minimal suprasellar extension (J.McK.) and in one patient (R.T.) with a large tumour which was extending laterally. The latter patient has since received treatment by external irradiation, which reduced her prolactin concentration from 160 pg/l to 88 pg/1 after 12 months. She and J.McK. have subsequently received bromocriptine as secondary therapy and both have resumed ovulatory menstrual cycles. Transfrontal surgery was required in 2 patients because of marked suprasellar extension of their tumours. One (B.H.) had a very high prolactin level of 1600 pg/l which fell to 530 pg/l after the operation and there was no return of menstrual cycles. The other patient (J.K.) had severe upward extension of her tumour which had caused internal hydrocephalus and papilloedema. Gonadotrophin secretion was impaired before and after surgery and despite a fall of her prolactin concentration to within the normal range she has not menstruated since the operation. Thus suprasellar extension of the pituitary tumour occurred in 4 of the 5 patients who did not ovulate after the operation and the only example of gonadotrophin deficiency directly attributable to surgery occurred in one of these

TF

Suprasellar extension

2

36

J.K.

TF

14

32

B.H. 0

+

0

TS = Transsphenoidal operation. (Mr R. A. Williams) T F = Transfrontal operation. (Mr J. Andrew) * Became pregnant after surgery t Responded to secondary therapy.

2

34

J.McK.

Minimal suprasellar extension Minimal suprasellar extension Suprasellar extension

0

TS

TS

TS TS

TS

TS TS TS

8

27

L.S.

0 0 0

0

+

32 23 29 27 23 26

No extension No extension No extension No extension No extension Lateral extension

Operation

+

AEG findings

1 Primary 5 6 7 2

Duration of amenorrhoea Galactorrhoea (years)

J.W. M.H. G.P. B.R. P.S. R.T.

Patient

~

25.4

1600

18.4

13.6

530

17.9

2.3

11.2 2.2 23.6 16.1 11.0 1 60

404 350 315 115 88 790 140

Post-op

Pre-op

Serum prolactin (pg/l)

TABLE 111 Details of patients treated surgically

Persisting amenorrhoea and galactorrhoea Persisting amenorrhoea (low gonadotrophins before operation)

Persisting amenorrhoea?

Ovulatory cycles* Ovulatory cycles Ovulatory cycles Ovulatory cycles Ovulatory cycles Persisting amenorrhoea and galactorrhoeat Panhypopituitarism

Response to surgery

VI

t3 P

E i?

80

k

R s

Ei%

P

d 0

Fdm;a

$

9 71

8 5

?!

5

5

246

FRANKS, JACOBS, HULL, STEELE AND NABARRO

patients (L.S.). The other patient who had a disappointing response to surgical treatment had a large tumour which was extending laterally. External irradiation External irradiation (4500 rads by 6 mev linear accelerator) was given as primary therapy to four patients. In one patient, prolactin levels fell over 12 months from 120 to 40 pg/l but the amenorrhoea and galactorrhoea remained unchanged. Subsequent treatment with bromocriptine (7.5 mg/day) as secondary therapy reduced the prolactin level to 17 pg/l but the amenorrhoea and galactorrhoea persisted. Three patients have recently completed external irradiation: in none had there been a fall of prolactin levels within 3 months and in all the amenorrhoea persisted. Treatment of these patients with bromocriptine is planned. Thyroxine The one patient in this series with hyperprolactinaemia and primary hypothyroidism was treated with thyroxine (1 50 pg/day).

After three weeks she was clinically and biochemically euthyroid, the galactorrhoea had subsided, and serum prolactin levels had fallen to within normal range. She has subsequently resumed normal ovulatory cycles. Acute response to bromocriptine The results of prolactin measurements every 30 minutes in 3 untreated patients from Group 1 are shown in Figure 2. There was marked episodic fluctuation of prolactin concentrations throughout the day, the minute-to-minute changes being most striking in the patient with the highest prolactin levels. There was no systematic increase of prolactin levels at night time. The response to oral bromocriptine is illustrated in Figures 3, 4 and 5. One tablet of 2.5 mg of bromocriptine given after supper produced a sustained fall of prolactin concentrations in two patients with basal levels around 20 pg/l (Fig. 3) but in the patient with a basal prolactin level of about 45 pg/1 the initial fall of prolactin was followed by escape at

81

s

t

1

L

10

12

4

MD

Pn

8

12

4

nN

AM

8

10

FIG.2 Basal serum prolactin concentrations measured every 30 minutes in three hyperprolactinaemic patients with normal X-ray appearance of the pituitary region(group 1). The results of J.Cr., referred to in the text, are indicated by the open circles. In each patient there were marked episodic fluctuations of prolactin concentrations.

MANAGEMENT OF HYPERPROLACTINAEMIC AMENORRHOEA

247

50 r

0

1

I

loo0 1200

I

1

0400

I

I

o800

I

I

1200

I

I

0400

I

1

I,

o800 lo00

FIG.3 Serum prolactin concentrations measured before and after one tablet of 2 . 5 mg of bromocriptine in three patients with normal X-ray appearance of the pituitary region (group 1). The time at which the tablet was swallowed is indicated by the vertical arrow. There was a prompt fall of prolactin concentrations which, in the patientswith the pretreatment levels of about 20 pg per litre, was sustained for up to 16 hours.

about 0400 hours. In one patient with a pituitary tumour there was a relatively poor and stepwise fall of prolactin from over 100 pg/l to about 45 pg/l 8 hours later (Fig. 4). In Figure 5, the prolactin levels after bromocriptine are expressed as a percentage of their mean (& ISD) pretreatment levels (expressed as 100 per cent) and plotted on the ordinate on a logarithmic scale. The half-time of the rate of removal of prolactin from the blood was about 4 hours in all patients. In 3 (including the one with a pituitary tumour) there was an initial rise of prolactin concentrations preceding the fall. Because of this delay in the fall of prolactin levels, a prolactin concentration of 50 per cent of the mean pretreatment level

was not seen till at least 5 hours after bromocriptine was given. Long term response to brornocriptine Primary therapy (32 patients). The results

of primary therapy with bromocriptine are shown in Table IV. All 32 patients (including 3 with pituitary tumours) developed ovulatory menstrual cycles and 14 conceived. One patient aborted spontaneously at 14 weeks gestation; amenorrhoea and galactorrhoea then recurred but she has taken bromocriptine again and is now 36 weeks pregnant (patient YU in Jacobs et al, 1976). The disparity between the rates of ovulation and conception in the remaining patients is accounted for by barrier methods

248

FRANKS, JACOBS, HULL, STEELE AND NABARRO

10

12

I

1

4

0

1

I

MD

12

4

MN

AM

0

10

FIG.4 Serum prolactin concentrations before and after one tablet of 2.5 mg of bromocriptine in a patient with an untreated pituitary tumour (from group 2). There was a slow and stepwise fall of prolactin concentrations over 12 hours but the levels did not fall to within the normal range.

of contraception in ten (including the three patients in Group 2) and blocked Fallopian tubes in two patients. One patient aged 42 has a unicornuate uterus, one is married to a man with oligospermia and one patient is single. The cause of the persisting infertility has not been established in three patients, one of whom has emigrated and been lost to follow-up.

Secondary therapy (8 patients). The results in three patients in Group 2 (R.T. and J.McK. treated surgically and one patient treated by external irradiation) have been described. Five patients in Group 3, whose details were reported by Jacobs et a1 (1976), have received secondary therapy with bromocriptine-two have ovulated and become pregnant. Anovular

TABLE IV Long term results of treatment with bromocripfine

Primary treatment Group 1 Group 2 Secondary treatment Group 2 Group 3 Totals

Number treated

Menstruation

Ovulation

29 3

29 3

29 3

3 5

2 2

2 2

40

36

36

* One patient has been pregnant twice.

Pregnancy

14 (+I)* 0 1

2

17 (+1)*

MANAGEMENT OF HYPERPROLACTINAEMIC AMENORRHOEA

0

L HOURS AFTER

8 12 BROMOCRIPTINE

16

FIG.5 Serum prolactin concentrations after bromocriptine, plotted on a logarithmic scale as a percentage of their mean pretreatment level (expressed as 100 per cent). The half time of the rate of fall of prolactin was about 4 hours.

menstrual cycles occurred in one patient with partial gonadotrophin deficiency but amenorrhoea has persisted in the remaining 2 patients who have post-operative panhypopituitarism. These results are summarized in Table IV.

DISCUSSION In treating these patients our objectives have been the restoration of fertility and normal ovulation cycles (and thereby the repair of oestrogen deficiency) and the elmination of galactorrhoea. The form of treatment selected for a particular patient has depended upon identification of the cause of hyperprolactinaemia. Where drugs were responsible, they were discontinued and the results have not been included here. If it is considered

249

inadvisable for a patient to stop the drug causing hyperprolactinaemia (for instance when phenothiazines are used for treating schizophrenia) concurrent treatment with bromocriptine is successful in lowering prolactin levels and apparently does not interfere with the psychotropic effect (Beumont et al, 1975). Hypothyroidism may cause hyperprolactinaemia (Bray and Jacobs, 1974) which can be reversed by treatment with thyroxine. It is important to emphasize that, since hypothyroidism that is mild enough to elude purely clinical diagnosis may be responsible, biochemical confirmation of the euthyroid state should always be sought in patients with hyperprolactinaemia. It has been known for several years that pregnancies which follow induction of ovulation in patients with untreated pituitary tumours may be complicated by sudden and dangerous enlargement of the pituitary tumour (Gemzell, 1975; Editorial, 1976), so in this series all patients have had careful radiological evaluation of the pituitary region. This should always include tomography of the fossa (Vezina and Sutton, 1974). In those with radiological abnormalities suggestive of an intrafossa tumour (Joplin e l al, 1975), an air encephalogram was performed to exclude an expanded but ‘empty’ sella (Kaufman et al, 1973) and to determine whether there was any upward extension of a pituitary tumour. If a pituitary tumour was present and the patient wished to conceive, initial therapy was attempted by direct treatment of the tumour, followed, if necessary, by therapy with bromocriptine. The intention was either radical cure of the prolactin producing pituitary tumour by surgery, or prevention by irradiation of pituitary expansion during a pregnancy subsequently induced by bromocriptine. Based on this approach, 36 patients with pituitary tumours have been evaluated before treatment and 17 have now completed at least primary therapy. Suprasellar extension occurred in 4 of the 36 patients with pituitary tumours and hyperprolactinaemia. In two patients (J.K. and J.McK.) the preoperative serum prolactin concentrations were not as high as they usually are in patients with prolactin secreting pituitary tumours (prolactinomas) (Child et al, 197%

250

FRANKS, JACOBS, HULL, STEELE AND NABARRO

and b; Jacobs, 1976). It is probable that these tumours were ‘non functioning’ and that the hyperprolactinaemia was secondary to impingement by the tumour on the hypothalamus or pituitary stalk, thus impairing the production or transport of the hypothalamic factor which normally restrains prolactin secretion. Nonetheless, the favourable response to suppression of prolactin secretion by bromocriptine in one of these patients (J.McK., Table 111) suggests that even such secondary hypersecretion of prolactin can cause the syndrome of hyperprolactinaemic amenorrhoea. In two patients suprasellar extension of the pituitary tumour was large enough to require operation by the transfrontal route. Complete removal of the tumour was not possible in one patient (B.H.) and in the other (J.K.), who had intracranial hypertension and papilloedema, there was evidence of gonadotrophin deficiency before the operation, so despite a fall of prolactin levels to normal reproductive function was not restored. In the other two patients the suprasellar extension was less marked and they were treated by transsphenoidal surgery. One patient (L.S.) developed hypopituitarism as a result of surgery and the other (J.McK.) required secondary therapy with bromocriptine because of an inadequate fall of the prolactin concentration. The one other patient with a disappointing response to surgery (R.T.) had a large laterally extending tumour. These results indicate that tumours which extend beyond the confines of the pituitary fossa are not uncommon in patients with hyperprolactinaemic amenorrhoea (14 per cent of those with pituitary tumours in this series) and that they carry a grave prognosis for future fertility. Parenthetically it is salutory to note that the patient with papilloedema had a two-year history of amenorrhoea. In contrast, in the five patients with no extension of the tumour beyond the sella, surgical treatment by the transsphenoidal route was extremely successful and these patients all developed ovulatory menstrual cycles, and one became pregnant, without additional treatment. The prevalence of hypopituitarism attributable to surgery (1 out of 9 patients) in this small series is similar to that in a much larger group of patients with acromegaly treated successfully

by transsphenoidal surgery, which was reported recently from the Middlesex Hospital (Williams et al, 1975). As in that series, complications of surgery occurred more frequently in the patients with the largest tumours. The results reported here may therefore be taken as representative of those to be expected when patients with pituitary tumours are treated by experienced surgeons. Clearly surgical treatment is mandatory in patients with large tumours with significant extension upwards towards the optic chiasma, laterally towards the cavernous sinus, or downwards into the sphenoid sinus. In patients with none of those indications for surgery an alternative treatment is internal (interstitial) (Child et al, 1975a and b) or external pituitary irradiation (Thorner et al, 1975). In this series, primary therapy with external irradiation was attempted in four patients. It is too soon to evaluate the results fully, but in the one patient whose primary treatment was completed more than a year ago, there has been only a gradual fall of serum prolactin concentrations from 120 to 40 pg/l. A similar delay in the fall of prolactin concentrations was seen in the patient given secondary therapy with external irradiation (R.T., Table 111). This slow response is consistent with the much longer experience of Reyes et a1 (1976) of external irradiation in patients with prolactinomas and it is similar to the rate of fall of growth hormone after pituitary irradiation in patients with acromegaly (Williams et al, 1975). Because of the very slow rate of fall of prolactin concentrations, our objective with external irradiation has not been the immediate reversal of the reproductive defect but rather the hope, admittedly not yet proven by controlled trial, of avoiding pituitary expansion in a subsequent pregnancy, for which induction of ovulation with secondary bromocriptine treatment is usually required (Thorner et al, 1975). Whether these young patients will develop hypopituitarism in the future is not yet known and careful long-term follow-up is clearly obligatory. In patients with pituitary tumours in whom there are neither neurosurgical indications for operative treatment nor a wish for conception the indications for treatment are less certain. Galactorrhoea may be present but it is less

MANAGEMENT OF HYPERPROLACTINAEMICAMENORRHOEA

common, and certainly less distressing, than the symptoms resulting from the severe oestrogen deficiency that characterizes the hyperprolactinaemic syndrome, and to which attention was drawn in the early descriptions of patients with amenorrhoea and galactorrhoea (Argonz and del Castillo, 1953; Forbes et al, 1954). In this series, 89 per cent of the patients who were tested had objective evidence of oestrogen deficiency, in that they failed to respond to progestogen challenge with a withdrawal bleed (Table 11) and we have previously reported the association of hyperprolactinaemic amenorrhoea with very low serum oestradiol concentrations (Jacobs et al, 1976). The resulting complaints of superficial dyspareunia, dryness of the vagina during intercourse and occasionally attacks of flushing can be readily reversed by effective suppression of prolactin secretion. As shown in Figure 4, even in patients with pituitary tumours prolactin levels can be suppressed by treatment with bromocriptine. However, when patients with pituitary tumours are treated with bromocriptine, pregnancy should be avoided by barrier methods of contraception. In our opinion, the combination of bromocriptine with this form of contraception is preferable to treatment with contraceptive steroids, since the oestrogenic component of the oral contraceptive has been shown to elevate serum prolactin levels (Abu-Fadil et al, 1976; Dericks-Tan and Taubert, 1976) and in animals treatment with oestrogen causes proliferation of pituitary lactotrophs (Lloyd et a/, 1973). There is also evidence from studies in animals that bromocriptine reduces mitotic activity of pituitary lactotrophs (Lloyd et al, 1975) and can inhibit growth of experimental prolactin-producing pituitary tumours (MacLeod and Leymeyer, 1973). Whether long-term treatment with bromocriptine will reduce the size of a human pituitary tumour or at least impair its rate of growth is a matter for further study. Bromocriptine has been used for treatment of selected patients in all three groups. The time course of the response of serum prolactin to oral bromocriptine is shown in Figures 3 to 5. The marked spontaneous fluctuation of prolactin levels in the untreated state (Fig. 2) is smoothed out and prolactin levels fall with a half-time of removal from the circulation of about four

25 1

hours. Since the true biological half-time of disappearance of prolactin from the circulation is about 20 minutes (Daughaday, 1974), these results indicate that, despite a fall of the levels to within the normal range, prolactin secretion had not stopped completely in these patients. In the patient with the highest pretreatment prolactin concentrations, a single tablet of 2. 5 mg of bromocriptine did not reduce the prolactin levels to normal. This finding explains why larger doses of bromocriptine are usually needed in patients with very high serum prolactin concentrations. Bromocriptine has so far been used as primary therapy in 29 out of a total of 45 patients with no radiological evidence of a pituitary tumour and in three patients with pituitary tumours. In these patients the criterion of a response to bromocriptine (the development of ovulatory menstrual cycles) was satisfied in all cases and 15 pregnancies occurred in 14 patients. Those not conceiving had, with three exceptions, adequate non-endocrine factors to explain their failure to conceive. The cause in two of the other patients is not yet known and the third was lost to follow-up. The favourable ovulation rate on treatment with bromocriptine is all the more remarkable when compared to the rate of response to clomiphene, since only one of the 29 patients tested ovulated normally with this treatment (Table 11). In patients with anovular infertility, the association of oestrogen deficiency, gonadotrophin concentrations that are not raised, and a failure to respond to clomiphene has usually been taken as an indication for induction of ovulation with gonadotrophin injections (Lunenfeld and Insler, 1974). Now that it is known that hypersecretion of prolactin may cause this pattern of reproductive defect even in the absence of galactorrhoea (Jacobs et al, 1976), treatment with bromocriptine has replaced the use of gonadotrophin injections in twothirds of the women requesting induction of ovulation (Jacobs et al, 1975). As indicated in Table IV, this treatment may be used as secondary therapy after unsuccessful treatment by surgery or by irradiation, or by both (R.T., Table 111) and three pregnancies have occurred in these patients (Table IV). The advantages of treatment with bromocriptine are

252

FRANKS, JACOBS, HULL, STEELE AND NABARRO

considerable : in contrast to gonadotrophin therapy the prevalence of multiple conceptions (2 per cent of 148 births; G. Kennedy, 1977, personal communication) and fetal abnormalities (2.7 per cent) in pregnancies induced with this compound is no higher than in the normal population and the need for expensive and time consuming clinical and biochemical monitoring is obviated. There have been few side-effects with the doses of bromocriptine needed to suppress hypersecretion of prolactin in these patients. Those that have occurred have usually been transient: they include nausea and vomiting, which can be avoided by taking the drug with food, and postural hypotension which can usually be avoided by taking the drug on retiring to bed (Leading Article, 1975). Occasionally constipation is troublesome but it usually responds to a simple change in diet. One patient developed a maculo-papular rash after a week’s treatment. The rash became worse when the dose of bromocriptine was raised and it disappeared within a week of stopping it. Reintroduction of bromocriptine one month later has not lead to recurrence of the rash and the patient has now been taking the drug for nine months without trouble. Two patients have had to stop their treatment with bromocriptine because of the development of severe headache; one of these (R.T.) had a pituitary tumour previously treated by surgery and external irradiation but the other had normal X-rays of the fossa. Four patients have failed to develop ovulatory menstrual cycles in response to secondary treatment with bromocriptine : two had complete and one partial gonadotrophin deficiency following previous pituitary ablative therapy and they will need treatment with gonadotrophins if induction of ovulation is required. However, in the other patient with a pituitary tumour who had received external irradiation as primary therapy, the failure to respond to bromocriptine (7.5 mg/day) was undoubtedly due to the persistence of elevated prolactin concentrations. This must therefore be regarded as a failure of therapy, though it seems probable that the patient will respond when the dose of bromocriptine is raised. The one disadvantage of treatment with

bromocriptine is that prolactin levels remain suppressed only while the drug is being taken. When treatment of two of our patients was intentionally stopped, in one case after uninterrupted medication for 18 months, serum prolactin concentrations rose to their pretreatment levels by the following day. Clinical relapse of amenorrhoea and galactorrhoea and recurrence of hyperprolactinaemia after discontinuing bromocriptine have been described by Thorner et af (1974). These observations may be considered to constitute a strong argument for definitive surgical extirpation of prolactin producing pituitary tumours, particularly as small (micro) tumours, which are not yet radiologically detectable, may be present in many patients with hyperprolactinaemic amenorrhoea (Hardy, 1973; Jacobs et af, 1976). The preferred approach would be Hardy’s (1973) technique of transsphenoidal micro-surgical selective removal of the adenoma, but this technique is not yet widely available. Our current policy of management of patients with hyperprolactinaemic amenorrhoea is therefore as follows. All patients have careful and detailed radiological assessment of the pituitary region. Drugs known to cause hyperprolactinaemia are either withdrawn or bromocriptine is given concurrently, and if hypothyroidism is present it is corrected. In patients with abnormal X-rays of the pituitary fossa (including tomography) an air encephalogram is performed. Transfrontal surgery is required if there is significant upward extension of a pituitary tumour. In patients with large tumours but without serious upward extension, and in those with other endocrine syndromes, for instance acromegaly (Franks et al, 1976), a transsphenoidal operation is advised. Patients with smaller tumours who wish to conceive are advised either to have external irradiation, followed usually by secondary treatment with bromocriptine, or a transsphenoidal surgical removal of the tumour. Results of surgical treatment are best in this type of patient, in that the five patients in our study who had tumours confined within the sella had definitive cures with no requirement for any post operative treatment. Primary therapy with bromocriptine is reserved for patients in whom careful

MANAGEMENT OF HYPERPROLACTINAEMIC AMENORRHOEA

radiological evaluation has produced no evidence of a pituitary tumour, and for those with tumours who have no wish to conceive, providing the air encephalogram has excluded upward extension of the tumour. It should be emphasized that repeat X-rays of the pituitary fossa should be obtained before each pregnancy is planned (Gemzell, 1975) and careful supervision, including assessment of the visual fields by experienced personnel, is needed throughout the period of antenatal care.

ACKNOWLEDGEMENTS The authors express their gratitude to Dr Graham Kennedy, of Sandoz Pharmaceuticals for financial support and the initial supplies of bromocriptine. We are grateful to Dr W. D. Odell, the National Pituitary Agency of the United States and the Medical Research Council for supplying reagents used in the radioimmunoassays. Excellent technical assistance was provided by Mr John Fox. Serum progesterone concentrations were measured in the Department of Chemical Pathology at St Mary’s Hospital Medical School and we express our gratitude to them. Finally, we would like to thank the gynaecologists and physicians of St Mary’s Hospital and The Middlesex Hospital for their kind permission to study cases under their care. REFERENCES Abu-Fadil, S., De Vane G., Siler, T. M., and Yen, S. C. C. (1976): Contraception, 13, 79. Argonz, J., and Del Castillo, E. B. (1953): Journal of Clinical Endocrinology and Metabolism, 13, 79. Beurnont, P., Bruwer, J., Pimstone, B., Vinik, A., and Utian, W. (1975): British Journal of Psychiatry, 126, 285. Bray, G. A., and Jacobs, H. S. (1974): Handbook of Physiology, Section 7, Vol 111. Edited by M. A. Greer and D. H. Solomon, p 413. Child, D. F., Gordon, H., Mashiter, K., and Joplin, G. F. (1975~):British Medical Journal, 4, 87. Child, D. F., Nader, S., Mashiter, K., Kzeld, M., Banks, L., and Russell Fraser, T. (19756): British Medical Journal, 1, 604. Daughaday, W. H. (1974): Text Book of Endocrinology. Edited by R. H. Williams. Saunders S. H., Philadelphia, p 31. Dericks-Tan, J. S. A., and Taubert, H. D. (1976): Contraception, 14, 1. Editorial (1976): Lancet, 1, 404.

253

Forbes, A. P., Henneman, P. H., Griswold, G. C., and Albright, F. (1954): Journal of Clinica[Endocrinology and Metabolism, 14, 265. Franks, S., and Jacobs, H. S . (1976): Prolactin and Human Reproduction. Edited by P. Crosignani, Academic Press, London. Franks, S., Jacobs, H. S., and Nabarro, J. D. N. (1976): Clinical Endocrinology, 5, 63. Franks, S., Murray, M. A. F., Jequier, A. M., Steele, S. J., Nabarro, J. D. N., and Jacobs, H. S. (1975): Clinical Endocrinology, 4, 597. Gemzell, C. (1975): American Journal of Obstetrics and Gynecology, 121, 311. Hardy, J. (1973): Diagnosis and Treatment of Pituitary Tumours. Edited by P. 0. Kohler and G. T. Ross. Int. Cong. Series No. 303, p 179. Hull, M. G. R., Franks, S., McNeilly, A. S., Murray, M. A. F., and Jacobs, H. S . (1975): Journal of Endocrinology, 67, 17. Hull, M. G. R.,Murray, M. A. F., Franks, S., Lieberman, B. A., and Jacobs, H. S . (1976): The Endocrine Function of the Human Ovary, Edited by M. Serio and V. H. James. Academic Press, London, p 245. Jacobs, H. S . (1976): New England Journal of Medicine, 295, 954. Jacobs, H. S., Franks, S., Murray, M. A. F., Hull, M. G. R., Steele, S. J., and Nabarro, J. D. N. (1976): Clinical Endocrinology, 5, 439. Jacobs, H. S., Hull, M. G. R.,Murray, M. A. F., and Franks, S. (1975): Hormone Research, 6, 268. Joplin, G. F., Jackson, R. A., Arnot, R. N., Burke, C. W., Doyle, F. H., Harsoulis, P., Lewis, D., Macerlean, P., Marshall, J. C., Van Noorden, S., and Russell Fraser, T. (1975) : Clinical Endocrinology, 4, 139. Kaufman, B., Pearson, 0. H., and Chamberlain, W. B. (1973): Diagnosis and Treatment of Pituitary Tumours. Edited by P. 0. Kohler and G. T. Ross. Int. Cong. Series No. 303, p 100. Leading Article (1975): British Medical Journal, 4, 667. Lloyd, H. M., Meares, J. D., and Jacobi, J. (1973): International Journal of Cancer, 11, 90. Lloyd, H. M., Meares, J. D., and Jacobi, J. (1975): Nature, London, 255, 497. Lunenfeld, B., and Insler, V. (1974): Clinical Endocrinology, 3, 223. Macleod, R. M., and Leymeyer, J. E. (1973): Cancer Research, 33, 849. Reyes, F. I., Gomez, F., and Faiman, C. (1976): Prolactin and Human Reproduction. Edited by P. Crosignani. Academic Press, London. Thorner, M. O . , Besser, G . M., Jones, A,, Dacie, J., and Jones, A. E. (1975): British Medical Journal, 4, 694. Thorner, M. O., McNeilly, A. S., Hagen, C., and Besser, G. M. (1974): British Medical Journal, 2, 419. Vezina, J. L., and Sutton, T. J. (1974): American Journal of Roentgenology, 120, 46. Williams, R. A,, Jacobs, H. S., Kurtz, A. B., Millar, J. G. B., Oakley, N. W., Spathis, G. S., Sulway, M. J., and Nabarro, J. D. N. (1975): Quarterly Journal of Medicine, 43, 79.

Management of hyperprolactinaemic amenorrhoea.

British Journal of Obstetrics I I I I and Gynaecology VOL 84 No. 4 NEW SERIES APRIL 1977 MANAGEMENT OF HYPERPROLACTINAEMIC AMENORRHOEA BY S. FR...
890KB Sizes 0 Downloads 0 Views