Short and long-term responses to metyrapone in the medical management of 91 patients with Cushing’s syndrome
J. A. Verhelst, P. J. Trainer, T. A. Howlett, L. Perry, Lesley H. Rees, A. B. Grossman, J. A. H. Wass and G. M. Besser
Departments of Endocrinology, Chemical Endocrinology and Reproductive Physiology, St Bartholomew’s Hospital, London, UK
(Received 19 December 1990; returned for revision 1 February 1991; finally revised 21 March 1991; accepted 16 April 1991)
Summary
OBJECTIVE To analyse the clinical and biochemical effects of metyrapone in the treatment of Cushing’s syndrome.
DESIGN An evaluation of the standard clinical practice at one Institution.
PATIENTS Ninety-one patients with Cushing’s syndrome: 57 pituitary-dependent Cushing’s disease, 10 adrenocorti- cal adenomas, six adrenocortical carcinomas and 18 ectopic ACTH syndrome.
MEASUREMENTS The acute response to metyrapone was assessed by measuring cortisol, 11-desoxycortisol and ACTH at 0, 1, 2, 3, 4 hours after a test dose of 750 mg of metyrapone. The longer-term effect of metyrapone was Judged by measuring serum cortisol at 0900, 1200, 1500, 1800, 2100 and sometimes 2400 h and calculating a mean. RESULTS A test dose of 750 mg of metyrapone decreased serum cortisol levels within 2 hours in all groups of patients and this effect was sustained at 4 hours. At the same time, serum 11-desoxycortisol levels increased in all patients, while plasma ACTH increased In patients with pituitary Cushing’s disease and the ectopic ACTH-syn- drome. Fifty-three patients with Cushing’s disease were followed on short-term metyrapone therapy (1 to 16 weeks) before other more definitive therapy. Their mean cortisol levels (median 654 nmol/l, range 408-2240) dropped to the target range of <400 nmol/l in 40 patients (75%) on a median metyrapone dose of 2250 mg/day (range 750- 6000). Metyrapone was given long term in 24 patients with Cushing’s disease who had been given pituitary irradia- tion, for a median of 27 months (range 3-140) with adequate control of hypercortisolaemia in 20 (83%). In 10 patients with adrenocortical adenomas and six with
Correspondence: Professor G. M. Besser, Department of Endocrinology, St Bartholomew’s Hospital, West Smithfield, London ECIA 7BE, UK.
adrenocortical carcinomas, metyrapone in a median dose of 1750 mg/day (range 750-6000) reduced their mean cortisol levels (median 847 nmol/l, range 408-2000) to <400 nmol/l in 13 patients (81%). In 18 patients with the ectopic ACTH-syndrome the ‘mean cortisol levels’, obtained from five or six samples on the test day (median 1023 nmol/l, range 823-6354) were reduced to < 400 nmol/l in 13 patients (70%), on a median dose of 4000 mg/day (range 1000-6000). Reduction of cortisol levels was clearly associated with clinical and biochemical improvement. The medication was well tolerated. Transient hypoadren- alism and hirsutism were unusual but were the most common side-effects.
CONCLUSIONS In our experience metyrapone remains a most useful agent for controlling cortisol levels in the management of Cushing’s syndrome of all types.
Transsphenoidal surgery has become the treatment of choice for pituitary-dependent Cushing’s disease (Mampalam et al., 1988; Melby, 1988). Pituitary irradiation or bilateral adrena- lectomy are usually performed when pitutiary surgery has failed. Surgery is the first-line treatment in adrenocortical tumours and the ectopic-ACTH syndrome when possible, while radiotherapy and/or chemotherapy are reserved for metastatic disease (Orth & Liddle, 1971). Ancillary medical management of hypercortisolaemia is indicated in moderate or severe cases (i) during the extended diagnostic procedures sometimes required for localization of the ACTH-source, (ii) in preparation for surgery, (iii) following radiotherapy, which can take between 1 and 10 years to become fully effective and (iv) as part of the palliative therapy in metastatic disease.
Over the years several drugs have been advocated for Cushing’s syndrome, but all have disadvantages. Amino- glutethimide is effective in lowering cortisol levels but its use has been limited by dizziness, ataxia, somnolence, rash and lethargy especially at higher doses and combination with metyrapone has been advocated to reduce side-effects (Child et al., 1976). o,p’-DDD in low doses is very efficient in inhibiting cortisol biosynthesis, but slow onset and hyper- cholesterolaemia are disadvantages (Luton et al., 1979). The introduction of ketoconazole has been hampered by hepato- toxicity (McCance et al., 1987), while trilostane has several side-effects and is ineffective in long-term treatment. Inhibi- tors of pituitary ACTH secretion such as bromocriptine (Hale et al., 1988), are only occasionally effective, while
cyproheptadine can cause obesity and psychosis (Krieger et al., 1975).
Since its introduction in 1958, metyrapone (2-methyl-1,2- bis-(3-pyridyl)-1-propanone) has been successfully used in patients with adrenocortical tumours and the ectopic-ACTH syndrome (Orth, 1972; Beardwell et al., 1981; Gormley et al., 1982; Thoren et al., 1985) and has been advocated in Cushing’s disease by ourselves (Jeffcoate et al., 1977) and others (Ross et al., 1979; Dickstein et al., 1986; Donckier et al., 1986). However, it has been stated that metyrapone is ineffective in Cushing’s disease without concomitant radio- therapy because the compensatory increase in ACTH over- comes the adrenal blockade and, furthermore, that side- effects as a result of mineralocorticoid and androgen excess limit its use (Orth, 1978). As this has not been our experience, we have now reviewed all the 91 patients treated by us with metyrapone for Cushing’s syndrome over the last 20 years.
Materials and methods
Patients
Of the 91 patients with Cushing’s syndrome 57 (aged 14-68; 43 female) had pituitary-dependent Cushing’s disease, 16 (aged 9-68; 12 female) had Cushing’s syndrome secondary to adrenocortical tumours (10 adenomas and six carcinomas) and 18 patients (aged 33-73; 10 female) had histologically proven ectopic-ACTH syndrome (four pulmonary and two thymic carcinoid tumours, four oat cell carcinomas of the lung and one of the gall bladder, three pancreatic tumours, two medullary thyroid carcinomas, one phaeochromo- cytoma, and one colonic carcinoma). Diagnosis was made on the basis of dynamic endocrine testing, radiology and/or venous sampling for ACTH as previously described (How- lett et al., 1986). In all cases the diagnosis has subsequently been confirmed by surgery, long-term follow-up, or at autopsy.
Metyrapone therapy
In an attempt to estimate the likely dose requirement at the start of metyrapone therapy we gave a 750 mg test dose of the drug at 0900 h and measured serum cortisol and, in a proportion of the patients, ACTH and/or 11-desoxycortisol levels, hourly for 4 hours.
Metyrapone was continued usually at doses of 500 mg, but occasionally 250 or 750 mg, given approximately 8 hourly with food. After 48 to 72 hours on metyrapone, circulating cortisol levels were determined, and in a proportion of patients ACTH and 11-desoxycortisol, in samples obtained at 0900, 1200, 1500, 1800, 2100 and sometimes 2400 h. From the average of these five or six samples the ‘mean level’ of each circulating hormone level was then calculated for each
patient. Subsequently, the metyrapone dose was adjusted between a minimum of 250 mg twice daily and a maximum dose of 1-5 g four times daily, with the aim of reducing the ‘mean cortisol level’ to a target range of 300-400 nmol/l. This range was taken to reflect a normal cortisol production rate and is associated with clinical remission (Jeffcoate et al., 1977). After correction of the dose, a new day curve of serum cortisol was performed 3 to 7 days later to allow further adjustment of metyrapone doses. This cycle was repeated until mean cortisol levels were in or near the target range or until the maximal tolerated dose of metyrapone was reached. Additional corticosteroid replacement therapy (block-and- replace regimen) was used only rarely, and then only when a rapid response was mandatory in malignancy or for the most severe cases, when maximal doses of metyrapone were begun immediately. The patient was then discharged and restudied 4 to 6 weeks later, when a new day curve was performed, since a progressive fall in levels was often seen over this time on a constant dose at the start of treatment. Further follow- up depended upon additional diagnostic procedures, choice of definitive therapy and evolution of the disease. Pituitary irradiation, which was our first choice of therapy for pituitary Cushing’s until 1982, was performed with the three-field technique as described earlier (Howlett et al., 1989).
Biochemical methods
Serum or plasma cortisol was measured fluorimetrically until 1982 (Mattingly et al., 1962), and thereafter by a specific radioimmunoassay (Cunnah et al., 1987). A close relation- ship was demonstrated between our fluorimetric and radioimmunoassays. Neither of these assays shows signifi- cant cross-reactivity with 11-desoxycortisol (fluorimetry nil, RIA less than 0-4%), the cortisol precursor before the metyrapone blockade. This is important since the high cross- reactivity of some commercially available RIA kits (quoted cross-reactivities: Medgenix, 16%; Serono 8-6%; Amersham 6%; Abbott 4.7%; Becton-Dickinson 3-5%) can give falsely high apparent RIA ‘cortisol’ levels because circulating 11- desoxycortisol levels are high as a result of the blockade by metyrapone and often exceed 1000 nmol/l. Plasma ACTH was measured by a specific radioimmunoassay, after extrac- tion on leached silica glass (Rees et al., 1972). Serum 11- desoxycortisol was measured after 1980 by a specific radioimmunoassay (Perry et al., 1982).
Clinical effects and adverse reactions
Changes in features of Cushing’s syndrome and adverse effects could be retrospectively assessed in 75 of the 91 patients (82%) before and after starting metyrapone treat- ment at one or more follow-up visits before definitive
| Circulating hormone levels | Diagnosis | Number | Results: median (range) | ||
|---|---|---|---|---|---|
| Start | 2 hours | 4 hours | |||
| Cortisol (nmol/l) | CD | 54 | 730 | 276 | 320 (67-1020) |
| (335-1700) | (67-713) | ||||
| Ectopic | 10 | 1150 (660-3700) | 480 (161-1300) | 325 (185-1250) | |
| Tumours | 10 | 770 (370-3200) | 262 (160-1600) | 274 (92-1050) | |
| ACTH* (ng/l) | CD | 38 | 73 (14-408) | 102 (13-438) | 125 (48-482) |
| Ectopic | 7 | 163 (100-675) | 199 (88-681) | 272 (120-1131) | |
| Tumours | 6 | <10 | <10 | <10 | |
| 11-Desoxycortisol (nmol/l) | CD | 14 | 27 (3-162) | 260 (14-476) | 289 (50-970) |
| Ectopic | 3 | 56 (46-60) | 340 (231-335) | 722 (419-800) | |
| Tumours | 3 | 56 | 121 | 167 | |
| (38-78) | (55-195) | (66-289) | |||
*ACTH ng/1 x 0.225=pmol/1.
treatment (surgery or radiotherapy). Psychiatric assessment was carried out in one-third of the patients by a standardized interview as described earlier (Jeffcoate et al., 1979). Potas- sium levels, lying blood pressure, fasting blood glucose levels, and/or oral glucose tolerance tests were done as part of the follow-up assessments.
Statistics
Individual or mean values for circulating cortisol, ACTH, 11-desoxycortisol and the metyrapone dose in each group of patients were expressed as the median and the total range, since the distribution of the data was non-Gaussian. For statistical analysis, either a Student’s t-test was used when logarithmic transformation of the data resulted in a normal distribution, or non-parametric tests (Wilcoxon). For clar- ity, biochemical and clinical data analyses were performed on grouped time intervals during which most patients had at least one day curve (range 1-6 curves).
Results
Acute effects of metyrapone
Data are available in 74 patients. The initial, 2 and 4-hour levels of circulating cortisol, ACTH and 11-desoxycortisol
after the test dose of metyrapone are shown in Table 1. Results were compared between Cushing’s disease, adreno- cortical tumours and the ectopic-ACTH syndrome. After metyrapone, cortisol levels dropped within 2 hours in all groups of patients. In Cushing’s disease, cortisol levels reached a minimum after 2 hours, followed by a small but signficant increase after 4 hours (P<0.01). In patients with the ectopic-ACTH syndrome and adrenocortical tumours, cortisol levels showed a similar sharp decrease 2 hours after metyrapone, in spite of higher basal levels. In contrast to Cushing’s disease, cortisol levels in these two groups further decreased by 4 hours. ACTH levels rose significantly both in Cushing’s disease and in the ectopic-ACTH syndrome (P<0.01 and P <0.01), while levels remained undetectable in adrenocortical tumours. 11-Desoxycortisol levels increased after metyrapone in all patients. The increase was less pronounced in adrenocortical tumours, but almost identical in ectopic-ACTH syndrome and Cushing’s disease.
Short-term therapy (1-16 weeks) in Cushing’s disease (Fig. 1)
Fifty-three patients with Cushing’s disease were given metyr- apone for a period of 1 to 16 weeks during diagnostic
800 r ( a )
Mean plasma ACTH (ng/1)
600
400
200
<10
3000
b)
2000
1600
Mean serum cortisol (nmol/L)
1200
800
400
<50
Pre -
48-72
1-3
4-6
2-3 Months
treatment Hours
Weeks
Time
procedures and/or in preparation for definitive therapy. As soon as 48 to 72 hours after starting metyrapone, the mean cortisol levels had dropped from a median of 654 to 440 nmol/l (P<0-001), a value only slightly above the target range. Over the next 1 to 3 weeks the metyrapone dose was
adjusted, resulting in a further decrease of mean cortisol levels to a median of 378 nmol/l. No escape was seen thereafter with a median mean cortisol level of 348 and 373 nmol/l (i.e. in the target range) after respectively 4-6 weeks and 2-3 months. The individual patient’s mean cortisol values were reduced to <400 nmol/l in 40 of the 53 patients (75%) after a median of 2 weeks (range 24 hours-12 weeks). Although the majority of the individual mean cortisol levels in these controlled patients were within the target range, the values were not always between 300 and 400 nmol/l, but rather distributed through and around this target range. The main reason for this is the spontaneous intermittent fluc- tations in cortisol hypersecretion in some patients, resulting in occasional and transient under or over-treatment on a fixed metyrapone dose, despite regular adjustments of the dose. In 13 of 53 patients (25%), we failed to suppress the mean cortisol levels below 400 nmol/l, although in two patients values dropped to between 400 and 450 nmol/l. In this group the mean cortisol levels were lower than the pretreatment levels (median 807 nmol/1, n=13, range 585- 1342) after 1-3 weeks (median 491 nmol/l, n=13, range 406-797) but increased again after 4-6 weeks (median 661 nmol/1, n = 10, range 413-1250) and 2-3 months (median 785 nmol/1, n=4, range 704-934).
A decrease in serum cortisol values to <400 nmol/l 4 hours after the test dose of 750 mg predicted a good response to metyrapone during short-term treatment in 88% of the patients, whereas inability to suppress acutely to levels <400 nmol/l was followed by satisfactory management in only 50% of the patients.
As shown in Fig. 1, plasma ACTH levels increased significantly during metyrapone treatment (P <0.001). This increase occurred predominantly during the first 4-6 weeks of treatment, after which no further significant increase was seen. An increase was seen in 76% of patients and no difference was seen between patients who were well con- trolled and those who were not. Concomitantly, 11-desoxy- cortisol levels significantly increased (P<0-001) during treatment from a median of 28 nmol/l (n= 15, range 14-280) to 374 nmol/l (n=11, range 72-1266) after 1-3 weeks, 502 nmol/l (n=11, range 163-1912) after 4-6 weeks and 518 nmol/l (n=5, range 140-2673) after 2-3 months (results not shown).
The median dose of metyrapone was increased from 2000 mg/day (n=46, range 500-6000) after dose adjustment 1-3 weeks after starting therapy, to 2250 mg/day (n=42, range 750-6000) after 6 weeks and was 2250 mg/day (n= 19, range 500-6000) after 2-3 months. A weak but significant correla- tion was found between the cortisol levels at diagnosis and the metyrapone dose at 4-6 weeks of therapy (n=42, correlation =0.56, P<0.001).
800
(0)
Mean plasma ACTH (ng/1)
600
400
. …
… .
200
…
..
$
$
$
:
<10
1200
( b )
Mean serum cortisol (nmol / L )
800
₹
400
$
…
…
O
< 50
Pre - treatment
0
3
6
1
2
3
4
5
6
Months
Years
Time
After between 1 and 16 weeks of therapy 20 patients proceeded to transsphenoidal surgery, 22 to radiotherapy and five to adrenalectomy. Six patients were continued on metyrapone alone.
Long-term treatment in Cushing’s disease (Fig. 2)
Twenty-four patients received metyrapone for a median of 27 months (range 3-140) after pituitary irradiation. Twenty had received metyrapone before radiotherapy for 1 to 3 months and the remaining four commenced metyrapone together with radiotherapy. At the start of radiotherapy, the majority of mean cortisol levels were under control or significantly better with metyrapone alone. During the first
year after radiotherapy the median mean cortisol level remained within the target range, 349 nmol/l after 3 months, 339 nmol/l after 6 months and 300 nmol/l after 1 year. Cortisol levels were well controlled in 20 of the 24 patients (83%). Of the four patients who were inadequately con- trolled, two were better controlled on combination treatment with o,p’-DDD, one proceeded to bilateral adrenalectomy and one was cured by transsphenoidal surgery after failed combination treatment with o,p’-DDD.
In the well controlled patients the median value of the mean cortisol level stayed within the target range over the following 1-6 years (Fig. 2). As in the short-term treatment, individual mean cortisol levels were at times above or under the target range because of the cyclical nature of Cushing’s
disease, but no patient required additional corticosteroid replacement therapy. After 6 years of treatment, six patients were still on metyrapone and nine had gone into remission after a median of 25 months (range 8-65). Two patients underwent bilateral adrenalectomy after 14-16 months because of hirsutism. One patient underwent transsphenoi- dal surgery after 2 years because of a growing macro- adenoma with high levels of ACTH (up to 5990 ng/1 (ng/1 × 0-225=pmol/l)). One patient was changed to a com- bination treatment of metyrapone with o,p’-DDD because of an escape of cortisol levels (538 nmol/l) on metyrapone alone. One patient with concomitant hyperprolactinaemia had a drop in ACTH on bromocriptine, after which metyrapone was changed to bromocriptine (case reported by Hale et al., 1988). Of the six patients who were treated longer than 6 years, one finally underwent adrenalectomy, one was changed to o,p’-DDD, and one to combination treatment with o,p’-DDD, in each case because of hirsutism after 86- 140 months of treatment. Two patients went into remission after 83-96 months and one died of a myocardial infarction.
As also shown in Fig. 2, ACTH levels after radiotherapy remained high on metyrapone for about one year after which a steady decrease was seen, reflecting the effects of radio- therapy. The metyrapone dose could be decreased from a median of 2000 mg/day (n=24, range 750-4000) at the start of therapy to 750 mg/day (n = 14, range 250-2250) after the second year. Reduction or cessation of metyrapone in a number of patients before complete remission of their Cushing’s disease resulted in an immediate rise in cortisol levels although usually not to levels as high as before treatment. In those patients in whom 11-desoxycortisol was measured on a number of occasions a high peak value could be seen within the first few months of treatment from a median basal level of 86 (n=8, range 14-280) to 936 nmol/1 (n=8, range 82-2384). As with ACTH, a decrease was seen after the first year following radiotherapy: median peak value 407 nmol/l (n=4, range 63-1440). Six patients were treated long term with metyrapone alone without radio- therapy because of uncertainty of the ACTH-source at the time and good control of the Cushing’s state on metyrapone (five patients) or patient’s preference (one patient). During short-term treatment mean cortisol levels dropped to < 400 nmol/l in all of them. Treatment became ineffective ultimately in three patients after 7-17 months. Two patients proceeded to bilateral adrenalectomy and one to transsphenoidal surgery and radiotherapy with ultimate cure. Three patients were controlled on metyrapone alone for 9, 60 and 173 months after which they went into remission. One of these patients relapsed 2 years later and had a bilateral adrenalec- tomy. The median dose of metyrapone was 2000 mg/day (range 500-3000).
2000
1600
Mean serum cortisol (nmol/L)
1200
800
400
<50
Pre-
48-72
2
3
4
treatment
Hours
Weeks
Time
Adrenocortical tumours (Fig. 3)
Metyrapone was given to 16 patients with adrenocortical tumours either before surgery or before therapy with o,p’- DDD. Mean cortisol levels after starting metyrapone fell from a median of 847 nmol/l into the target range (340 nmol/l) after 48 to 72 and no escape was seen thereafter. Metyrapone was combined with replacement prednisolone in eight patients because of a deliberate policy of maximum reduction of cortisol levels. All but three patients (cortisol 630-850 nmol/l) obtained mean cortisol levels of <400 nmol/l (81%). After 2-8 weeks, 12 patients proceeded to adrenalectomy, three to therapy with o,p’-DDD, two in combination with metyrapone. One patient with dissemi- nated adrenocortical carcinoma died after 3 weeks. ACTH remained undetectable or at the lower limit of normal in all patients. The median metyrapone dose was 1750 mg/day (range 750-6000).
The ectopic-ACTH syndrome (Fig. 4)
Twenty patients with the ectopic-ACTH syndrome received metyrapone either during diagnostic work-up while search- ing for the ACTH source in occult disease (11 patients) or as
7000
6000
5000
4000
Mean serum cortisol (nmol/L)
3000
2000
1200
800
400
<50
Pre -
48-72
1-3
I
3
6
9
1
1/2
treatment Hours Weeks
Months
Years
Time
palliative treatment together with chemotherapy in metasta- tic disease (seven patients). On metyrapone, mean cortisol levels dropped from a median of 1023 nmol/l before therapy, to 710, 405 and 345 nmol/l after respectively 48-72 hours, 1-3 weeks and 1 month of treatment. Therapy thereafter continued to be effective, without any evident escape. In 13 of the 18 patients (70%) mean cortisol levels dropped to < 400 nmol/l. Of the five patients who were insufficiently con- trolled, two proceeded to bilateral adrenalectomy, one to combination treatment with o,p’-DDD, and two patients died soon after diagnosis. The other 13 patients continued metyrapone until curative surgery (seven, of whom six had occult disease) 1 to 20 months after starting therapy, or death (five, all of whom had overt disease) after 1 week to 44 months (median 6 months). One patient is still on therapy. As in some patients with adrenal tumours, six patients were deliberately overtreated with high doses of metyrapone in order to reduce cortisol production as much as possible, and
then received replacement prednisolone therapy. In contrast to the acute study ACTH levels did not rise during long-term metyrapone treatment, and showed considerable fluctation. The median ACTH level before therapy was 224 ng/1 (n = 18, range 37-2341) which is much higher than in Cushing’s disease (72 ng/1, n=49, range 20-387; ng/1 × 0-225=pmol/l). The metyrapone dose in patients with the ectopic-ACTH syndrome was also higher than in Cushing’s disease: median dose of 4000 mg/day (range 1000-6000) vs 2250 mg/day (500-6000).
Clinical and parallel biochemical effects
Lowering serum cortisol was associated with clinical im- provement in the majority of patients. Symptoms of bloated face and muscle weakness improved in 89 and 79% of patients respectively, occurring in a few days to several weeks. Psychiatric disturbances (depression, mood changes)
improved in 73% of patients in whom it was seen at presentation (73% Cushing’s disease, 60% ectopic-ACTH syndrome and 75% adrenocortical tumours). The three patients who had severe psychosis on admission showed great improvement in cognitive status one week after starting metyrapone. Despite control of cortisol levels, some milder degree of psychological abnormality persisted in many patients for several months. Hirsutism and/or acne was present in 92% of women before therapy and improved in only five (9%).
Potassium levels were low (<3.5 nmol/l) at diagnosis in 19% of patients with Cushing’s disease, 100% of those with the ectopic-ACTH syndrome and 44% of adrenocortical tumours. In all patients potassium levels rapidly normalized during metyrapone therapy. Hypertension was found in 81% of patients before treatment and improved in 73% on metyrapone. Glucose intolerance or diabetes mellitus was seen in 47% of patients with Cushing’s disease, 85% of ectopics and 43% of adrenocortical tumours. It improved in 82% of patients resulting in better control on diet alone, oral antidiabetic treatment or less insulin.
Adverse reactions Transient hypoadrenalism with symptoms such as nausea, vomiting, hypoglycaemia and postural hypotension associated with cortisol levels <200 nmol/l occurred in 10 of the 79 patients (13%) maintained on metyrapone without routine corticosteroid cover. Adreno- cortical insufficiency was precipitated by an infection (four patients) or occurred as a result of overtreatment (six patients). It was managed by reducing metyrapone or stopping the drug for a few days, while giving corticosteroid substitution. Transient mild hypokalaemia (range 2.9-3-3 mmol/l) was seen in five patients who had presented without hypokalaemia and responsed well to potassium supple- ments, amiloride or triamterene. Because hypokalaemia can occur spontaneously in Cushing’s syndrome, a relationship with metyrapone remains uncertain. Dangerous hyper- kalaemia (6-1 mmol/l) occurred once when an initially hypokalaemic patient on potassium supplements became hypoadrenal. Transient worsening of oedema was seen in six patients and was treated by short courses of diuretics. A rash occurred in three patients during the first weeks of therapy but metyrapone could be continued without further prob- lems. Twelve patients complained of dizziness or were light- headed on metyrapone. Minor gastrointestinal complaints, principally nausea, were reported but this occurred in the context of suboptimal circulating cortisol levels, except in four patients who felt nauseated despite normal circulating cortisol concentrations.
Hirsutism and/or acne was noted in nine of 16 women treated long term; however, it was already present in six of
these women before starting therapy. Metyrapone was associated with an increase in testosterone levels in women from a median of 2-2 nmol/l (n=30, range 0-8-4-4 nmol/l; normal for women 0-5-3 and for men 9-35) to a median peak of 6-7 nmol/l (n=18, range 2-7-17-2)
Discussion
The high circulating levels of cortisol in Cushing’s syndrome are usually associated with severe physical and emotional disturbances, which usually improve rapidly once cortisol levels fall. We have found that metyrapone is a highly effective drug for reducing cortisol levels in all causes of Cushing’s syndrome and escape is very rare once control is achieved. This contradicts an earlier report (Orth, 1978). The higher drive by compensatory increase of ACTH in Cush- ing’s disease does not overcome the blockade of 118- hydroxylase and cortisol production can be effectively suppressed in 75% of patients. The metyrapone test dose gave some indication of the initial dose requirement, but failed to differentiate clearly patients responding to therapy from the others.
Long-term treatment of Cushing’s disease with metyra- pone after pituitary radiotherapy is very effective and this experience supports the prediction from our initial report (Jeffcoate et al., 1977) and that of others after shorter periods of treatment (Ross et al., 1979). ACTH levels gradually decreased after radiotherapy making metyrapone treatment progressively easier to manage with lower doses until remission is reached. If the effect of radiotherapy is consi- dered in the whole group of patients, both the patients off and on metyrapone therapy, the decrease in ACTH levels is even more pronounced because the patients with the lowest ACTH values are usually in remission and off metyrapone therapy (Howlett et al., 1989). We only occasionally used metyrapone long term without any additional therapy in patients with Cushing’s disease. All these patients achieved cortisol levels < 400 nmol/l on metyrapone, although some escaped after 7-17 months. This therapy may be indicated alone in selected cases when, for example, the ACTH source is as yet unknown (Howlett et al., 1986), the condition is mild, or if radiotherapy and surgery are unwanted or associated with a high risk. Excellent control of Cushing’s disease with metyrapone has been achieved by others in this way in a child during puberty (Dickstein et al., 1986), and in an elderly patient (Donckier et al., 1986).
The effects of metyrapone in the ectopic-ACTH syndrome and adrenocortical tumours confirms earlier reports by documenting the excellent and fast control that can be obtained (Orth & Liddle, 1971; Orth, 1972; Beardwell et al., 1981; Gormley et al., 1982; Thoren et al., 1985). Since some
of these patients were not controlled with metyrapone despite high doses and undetectable plasma ACTH levels, we suggest that other mechanisms must exist other than increased ACTH-drive to account for incomplete blockade.
In our experience metyrapone and o,p’-DDD are the most useful adrenolytic agents in Cushing’s syndrome (Trainer et al., 1989). The advantage of meytrapone over o,p’-DDD, which takes 4-6 weeks to become fully active (Luton et al., 1979), is the immediate blockade that can be achieved and makes metyrapone particularly useful in obtaining a fast remission in severely ill and/or psychotic patients. A similar effect can be obtained with ketoconazole (Sonino et al., 1985) but this can cause hepatotoxicity, which seems to occur more often in Cushing’s syndrome than in patients with monilial infections (McCance et al., 1987).
We favour preparation of patients for surgery by a short course (usually 6 weeks) of metyrapone to reduce physical and emotional morbidity and post-opeative complications. Metyrapone therapy carries a risk of acute transient hypoad- renalism if the dose is too high. Similar effects have been described with ketoconazole (McCance et al., 1987). Preven- tive corticosteroid substitution therapy could reduce this risk but then necessitates higher doses of metyrapone to reduce hypercortisolaemia to much lower levels and increases the number of drugs to be taken by the patient, and is more expensive. We do not use this ‘block-and-replace’ regimen in Cushing’s disease.
Acute stress, such as infection, precipitated Addisonian crises on four occasions and patients on adrenolytic therapy should therefore carry a steroid emergency card. Hirsutism and/or acne as a result of accumulation of androgen precursors from the adrenals, is a well recognized problem with metyrapone especially during long-term treatment. Addition of cyproterone acetate produces improvement in some but not all patients. For this reason we now prefer to use o,p’-DDD alone or in combination with low doses of metyrapone if long-term treatment is expected in women. o,p’-DDD, however, increases cholesterol levels (Luton et al., 1979), and this problem has yet to be resolved. Persistent hypertension, oedema and hypokalaemia might be expected often, but occurred in only a few patients; it could be due to mineralocorticoid excess by accumulation of desoxycortico- sterone secondary to 11B-hydroxylase inhibition. These symptoms are, however, difficult to distinguish from the symptoms of Cushing’s syndrome itself and, moreover, they generally improve rather than worsen during continued therapy. Dizziness, gastrointestinal complaints and rash were never severe enough to necesitate discontinuation of metyrapone.
In summary, metyrapone is a valuable drug in the medical management of Cushing’s syndrome of all types, especially
when a rapid remission in severe cases is needed. It can be used long term if required. Significant side-effects are unusual and mild: most prominent is hirsutism during long- term treatment in women, and adrenocortical insufficiency due to intercurrent infections or over-treatment.
References
Beardwell, C.G., Adamson, A.R. & Shalet, S.M. (1981) Prolonged remission in florid Cushing’s syndrome following metyrapone treatment. Clinical Endocrinology, 14, 485-497.
Child, D.F., Burke, C.W., Burley, D.M., Rees, L.H. & Fraser, T.R. (1976) Drug control of Cushing’s syndrome. Combined amino- glutethimide and metyrapone therapy. Acta Endocrinologica, 82, 330-341.
Cunnah, C., Jessop, D.S., Besser, G.M. & Rees, L.H. (1987) Measurement of circulating corticotrophin-releasing factor in man. Journal of Endocrinology, 113, 123-131.
Dickstein, G., Lahav, M., Shen-Orr, Z., Edoute, Y. & Barzilai, D. (1986) Primary therapy for Cushing’s disease with metyrapone. Journal of the American Medical Association, 255, 1167-1169.
Donckier, J., Burrin, J.M., Ramsey, I.D. & Joplin, G.F. (1986) Successful control of Cushing’s disease in the elderly with long term metyrapone. Postgraduate Medical Journal, 62, 727-730.
Glaser, B., Kahana, L., Elias, V. & Shainfield, M. (1984) Sodium valproate and metyrapone for pituitary-dependent Cushing’s disease. Lancet, ii, 640.
Gormley, M.J.J., Hadden, D.R., Kennedy, T.L., Montgomery, D.A.D., Murnaghan, G.A. & Sheridan, B. (1982) Cushing’s syndrome in pregnancy-treatment with metyrapone. Clinical Endocrinology, 16, 283-293.
Hale, A.C., Coates, P.J., Doniach, I., Howlett, T.A., Grossman, A., Rees, L.H. & Besser, G.M. (1988) A bromocriptine responsive corticotroph adenoma secreting a-MSH in a patient with Cush- ing’s disease. Clinical Endocrinology, 28, 215-223.
Howlett, T.A., Drury, P.L., Perry, L., Doniach, I., Rees, L.H. & Besser, G.M. (1986) Diagnosis and management of ACTH- dependent Cushing’s syndrome: Comparison of the features in ectopic and pituitary ACTH production. Clinical Endocinology, 24, 699-713.
Howlett, T.A., Plowman, P.N., Wass, J.A.H., Rees, L.H., Jones, A.E. & Besser, G.M. (1989) Megavoltage pituitary irradiation in the management of Cushing’s disease and Nelson’s syndrome: Long-term follow-up. Clinical Endocrinology, 31, 309-323.
Jeffcoate, W.J., Rees, L.H., Tomlin, S., Jones, A.E., Edwards, C.R.W. & Besser, G.M. (1977) Metyrapone in long-term manage- ment of Cushing’s disease. British Medical Journal, ii, 215-217.
Jeffcoate, W.J., Silverstone, J.T., Edwards, C.R.W. & Besser, G.M. (1979) Psychiatric manifestations of Cushing’s syndrome: Re- sponse to lowering of plasma cortisol. Quarterly Journal of Medicine, 191, 465-472.
Krieger, D.T., Amorosa, L. & Linick, F. (1975) Cyproheptadine- induced remission of Cushing’s disease. New England Journal of Medicine, 293, 893-895.
Luton, J.P., Mahoudeau, J.A., Bouchard, Ph., Thieblot, Ph., Hautecouverture, M., Simon, D., Laudat, M.H., Toutiou, Y. & Bricaire, H. (1979) Treatment of Cushing’s disease by o.p’-DDD. Survey of 62 cases. New England Journal of Medicine, 300, 459-464.
Mampalam, T.J., Tyrrell, J.B. & Wilson, C.B. (1988) Transsphenoi- dal microsurgery for Cushing’s disease: a report of 216 cases. Annals of Internal Medicine, 109, 487-493.
Mattingly, D. (1962) A simple fluorimetric method for the estima- tion of free 11-hydroxycorticosteroids in human plasma. Journal of Clinical Pathology, 15, 374-379.
McCance, D.R., Hadden, D.R., Kennedy, L., Sheridan, B. & Atkinson, A.B. (1987) Clinical experience with ketoconazole as a therapy for patients with Cushing’s syndrome. Clinical Endocrino- logy, 27, 593-599.
Melby, J.C. (1988) Therapy for Cushing’s disease: A consensus for pituitary microsurgery. Annals of Internal Medicine, 109, 445-446.
Orth, D.N. & Liddle, G.W. (1971) Results of treatment of 108 patients with Cushing’s syndrome. New England Journal of Medicine, 285, 243-247.
Orth, D.N. (1972) Results of treatment in Cushing’s syndrome. In Cushing’s syndrome. Diagnosis and Treatment (eds C. Binder & P. Hall), pp. 148-152, Heinemann, London.
Orth, D.N. (1978) Metyrapone is useful only as adjunctive therapy in Cushing’s disease. Annals of Internal Medicine, 89, 128-130.
Perry, L.A., Al-Dujali, E.A.S. & Edwards, C.R.W. (1982) A direct radioimmunoassay for 11-deoxycortisol. Steroids, 39, 115-128.
Rees, L.H., Cook, D.M., Kendall, J.W., Allen, C.F., Kramer, R.M. & Ratcliffe, J.G. (1972) A radioimmunoassay for rat plasma ACTH. Endocrinology, 89, 254-261.
Ross, W.M., Evered, D.C., Hunter, P., Benham, M., Cook, D. & Hall, R. (1979) Treatment of Cushing’s disease with adrenal blocking drugs and megavoltage therapy to the pituitary. Clinical Radiology, 30, 149-153.
Sonino, M., Boscaro, M., Merola, G. & Mantero, F. (1985) Prolonged treatment of Cushing’s disease by ketoconazole. Journal of Clinical Endocrinology and Metabolism, 61, 718-722.
Thoren, M., Adamson, U. & Sjoberg, H.E. (1985) Aminogluthemide and metyrapone in the management of Cushing’s syndrome. Acta Endocrinologica, 109, 451-457.
Trainer, P.J., Howlett, T.A. & Besser, G.M. (1989) The medical management of Cushing’s sydrome using drugs acting on the adrenal cortex. In Recent Advances in Basic and Clinical Endocri- nology (eds F. Casanueva & C. Dieguez), pp. 341-349, Excerpta Medica, Oxford.