Clinical Endocrinology (1999) 51, 61–66

61q 1999 Blackwell Science Ltd

Growth hormone status following treatment forCushing’s syndrome

N. R. Hughes, C. A. Lissett and S. M. ShaletDepartment of Endocrinology, Christie Hospital,Manchester, UK.

(Received 24 September 1998; returned for revision13 November 1998; finally revised 16 December 1998;accepted 1 February 1999)

Summary

OBJECTIVE Both pituitary surgery and radiotherapyfor Cushing’s disease can lead to growth hormone(GH) deficiency. Studies to date have, however,described the incidence of impaired GH secretionand not the incidence of severe GH deficiency follow-ing treatment of Cushing’s disease. Furthermore,following cure of Cushing’s disease and resolutionof hypercortisolaemia, recovery of GH secretorystatus is seen, thus creating uncertainty as to thepersistence of any documented GH deficiency. Thisstudy has two aims; to determine the incidence ofsevere persistent GH deficiency following treatmentof Cushing’s disease and to assess the time scale ofany recovery of GH secretory status followingsurgical cure of Cushing’s disease.DESIGN AND PATIENTS The case notes of 37 patientseither cured or in clinical and biochemical remissionfollowing treatment for Cushing’s syndrome werereviewed to determine the incidence of severe GHdeficiency. Of 34 patients with Cushing’s disease, 20were treated by pituitary surgery, and 14 with radio-therapy. Three patients with adrenal adenomasunderwent unilateral adrenalectomy.MEASUREMENTS GH secretory status was assessedby provocative testing using an insulin tolerance test(ITT, 85% of all tests), glucagon stimulation test (GST)or arginine stimulation test (AST).RESULTS Thirty-six percent (5/14) of radiotherapytreated patients demonstrated severe GH deficiencyat a mean time of 99 months following remission.Fifty-nine percent (10/17) of surgically treated patientsassessed in the two years following remissiondemonstrated severe GH deficiency, whilst only 22%

(2/9) of patients assessed beyond two years followingremission demonstrated severe GH deficiency. Thislatter cohort is biased, with patients in whom severeGH deficiency had been demonstrated on earlier testsbeing over-represented. It is more accurate to esti-mate the incidence of persistent severe GH deficiencyfollowing surgically induced remission of Cushing’sdisease by incorporating data from patients in whomoriginal testing demonstrated adequate GH reserve.Collating such data, 13% (2/15) of patients had persis-tent severe GH deficiency. Across all time periods fivesurgically treated patients demonstrated recovery ofGH secretory status over a median time course of19 months.

In the surgically treated cohort, seven (35%)patients had anterior pituitary hormone deficitsother than GH deficiency: 14% (2/14) of patients withnormal GH secretory status at the last assessment,83% (5/6) of patients with severe GHD at the lastassessment. Of the 5 patients who demonstratedrecovery of GH secretory status 40% (2) hadadditional anterior pituitary hormone deficits.

Within the radiotherapy treated cohort 14% (2/14) ofpatients demonstrated additional anterior pituitaryhormone deficits: 11% (1/9) of patients with normalGH secretory status and 20% (1/5) of patients withsevere GH deficiency.

None of the patients with adrenal adenomas treatedby unilateral adrenalectomy demonstrated anyabnormality of GH secretory statusCONCLUSIONS The incidence of severe persistentGH deficiency following surgically induced or radio-therapy induced remission of Cushing’s disease islower than has been suggested by previous studies,although these latter studies have assessed GH insuf-ficiency and not severe GH deficiency. In the presenceof additional pituitary hormone deficits severe GHD iscommon and is likely to be persistent. Recovery of GHsecretory status is seen in a high proportion ofpatients reassessed, at a median time of 19 monthsfollowing surgically induced remission of Cushing’sdisease. Thus, we recommend that definitive assess-ment of GH secretory status is delayed for at least twoyears following surgical cure of Cushing’s disease.This has important implications for patients in whomGH replacement therapy is being considered.

Correspondence: Professor S. M. Shalet, Department ofEndocrinology Christie Hospital NHS Trust Wilmslow RoadManchester, M20 4BX, UK. Fax:þ44 (0)161 446 3772

Treatment options for Cushing’s disease include pituitarysurgery and radiotherapy, both of which may lead to GHdeficiency. Over the last decade it has been increasinglyrecognized that GH deficiency in adults is associated withchanges in various biological parameters, which include lipidprofile, insulin status, bone mineral density, body composition,physical performance and quality of life measures (McGauley,1989; Binnertset al., 1992; Cuneoet al., 1992); furthermore,GH replacement therapy can reverse many of these changes(Jorgensenet al., 1989; Salomonet al., 1989). These latterstudies and recent international recommendations, suggest thata peak GH response of less than 9 mU/l following an ITT beused to define severe GH deficiency (Growth HormoneResearch Society, 1998) and at present GH replacementtherapy is only recommended in patients who fulfil thesecriteria. However, previous studies reporting impaired GHstatus following treatment for Cushing’s disease have adoptedthe less stringent definition borrowed from the paediatricliterature of a peak GH response less than 20 mU/l to an ITT(Kuwayamaet al., 1981; Sempleet al., 1984; Littley et al.,1990). Thus, the incidence of severe GH deficiency followingtreatment of Cushing’s disease is unknown.

The situation is further complicated by evidence that GHsecretion is suppressed in the hypercortisolaemic state (Frantz& Rabkin, 1964) and that this suppression may continue for upto a year after resolution of hypercortisolaemia (Magiakouet al., 1994); although a single case of return of normal GHstatus has been described at least 21 months following pituitarysurgery for Cushing’s disease (Sempleet al., 1984). Previousstudies examining the incidence of GH deficiency followingtreatment of Cushing’s disease have assessed GH status in theshort term following pituitary surgery (Tyrrellet al., 1977;Kuwayamaet al., 1981; Burkeet al., 1990; Magiakouet al.,1994). Thus, patients in these studies may be suffering fromtemporary suppression of GH secretion as a result ofhypercortisolaemia rather than persistent surgically inducedGH deficiency, leading to an overestimation of severe GHdeficiency.

The aims of this study are twofold. Firstly, to determine theincidence of severe persistent GH deficiency followingtreatment of Cushing’s disease. Secondly, to assess the timescale of recovery from GH suppression following surgical cureof Cushing’s disease.

Patients and methods

The notes of 37 patients (31 female, 6 male) with Cushing’ssyndrome who presented between 1975 and 1996 werereviewed. These comprised 34 patients with Cushing’s diseaseand three patients with adrenal adenomas. Ages ranged from 12to 63 (mean 31·2) years.

Cushing’s syndrome during this period was diagnosed by thepresence of characteristic clinical features and confirmed byelevated urinary free cortisol on two occasions, loss of diurnalrhythm of serum cortisol and a midnight serum cortisol>300 nmol/l. A diagnosis of pituitary-dependent disease wasdemonstrated by a combination of the following; a mildlyraised or normal ACTH level, suppression of corticosteroidsecretion by 50% or more after a 48-h high dose (8 mg/24 h)dexamethasone suppression test, the presence of a pituitaryadenoma on CT or MRI scan, the absence of an adrenal tumouron CT scan and bilateral inferior petrosal sinus sampling withCRH induced stimulation of ACTH.

Treatment and remission

Of the 34 patients with Cushing’s disease, 14 received pituitaryirradiation as their primary treatment and 20 pituitary surgery.All patients underwent assessment of GH secretory statusbefore receiving any other therapy which may affect thehypothalamic/pituitary axis. To study the recovery of GHrelease following surgically induced remission of Cushing’sdisease, the surgically treated cohort was divided into thoseassessed within 24 months of remission and those assessedmore than 24 months following remission.

Pituitary irradiation was administered via a linear accelerator(8 MeV) employing two opposed lateral wedge fields and asingle frontal field. The total dose delivered was 2000 cGys ineight fractions. This is a smaller dose than that employed bymany other centres. Following irradiation, patients were treatedwith metyrapone (0·5–4 g) and/or aminoglutethimide (0·75–1·5 mg) for periods ranging from 3 weeks to 152 months.Following irradiation, remission was defined as a fall of urinaryfree cortisol to within the normal range. Medical therapy wasdiscontinued when biochemical remission had been achieved.

Pituitary surgery was performed by two different units,Professor G. Teasdale at Southern General Hospital, Glasgow(n¼ 18) and Mr C. B. T. Adams at Radcliffe Infirmary, Oxford(n¼ 2). Following surgery, pathological confirmation oftumour removal was obtained. Post-operatively, biochemicalconfirmation of remission was taken as a reduction in urinaryfree cortisol secretion to within the normal range and a low/normal serum cortisol concentration.

Three patients with adrenal adenomas underwent unilateraladrenalectomy.

Endocrinological follow-up

Long-term follow-up was undertaken in all patients withregular assessment of endocrinological status. Remission wasconfirmed by review of clinical status and assessment of 24 hurinary free cortisol concentration. Residual pituitary function

62 N. R. Hughes et al.

q 1999 Blackwell Science Ltd,Clinical Endocrinology, 51, 61–66

was assessed using an ITT (soluble insulin 0·2 U/kg i.v.), a GST(glucagon 1 mg, i.m.) or an AST (20 g/m2 arginine i.v. as a 20%solution over 30 minutes). Any cortisol replacement wasomitted the evening before and on the morning of testing inthe majority of patients: In four surgically treated patients andthree radiotherapy treated patients who has undergone bilateraladrenalectomies, cortisol replacement therapy was continuedduring the tests.

The status of other anterior pituitary hormones was assessedconventionally. At each assessment, basal blood specimens

were taken at 0900 h after an overnight fast for prolactin,thyroxine, TSH, FSH, LH, testosterone or oestradiol and sexhormone binding globulin (SHBG) concentrations. Patientswho had undergone bilateral adrenalectomy did not haveassessments of ACTH secretion

In line with a recent consensus statement (Thorneret al.,1995), severe GH deficiency was defined as a peak level of lessthan 9 mU/l an ITT. To exclude the effects of cortisol excessupon GH status, a diagnosis of severe GH deficiency was onlysought in patients without evidence of hypercortisolaemia.

GH status in treated Cushing’s disease 63

q 1999 Blackwell Science Ltd,Clinical Endocrinology, 51, 61–66

Table 1 Endocrine results following primary treatment for Cushing’s disease

Time elapsed Pituitary deficienciesPatient Primary postremission GH peakNo treatment (months) response GH FSH/LH ACTH TSH ADH

1 tsp 43 2·4 y n N/A n n2 tsp 19 3·3 y n y n n3 tsp 21 2 y n y n n4 tsp 36 19 n n n n n5 tsp 5 23 n n n n n6 tsp 30 10·8 n n n n n7 tsp 17 19 n n N/A n y8 tsp 14 18 n n n n n9 tsp 7 48 n n n n n

10 tsp 18 17 n n n n n11 tsp 63 53 n n n n n12 tsp 113 66 n n n n n13 tsp 19 12·6 n n n n n14 tsp 8 1 y y y n n15 tsp 71 35 n n n n n16 tsp 11 23·1 n n n n n17 tsp 84 15 n n y y y18 tsp 95 36 n n y n n19 tsp 4 7·2 y y N/A n n20 tsp 48 1 y y N/A y y21 XRT 87 12 n n n n n22 XRT 179 7·3 y y y y n23 XRT 120 48 n n n n n24 XRT 51 1·8 y n n n n25 XRT 99 14·6 n n n n n26 XRT 4 6 y n N/A n n27 XRT 103 19 n n N/A n n28 XRT 83 3·1 y n n n n29 XRT 46 1·5 y n n n n30 XRT 99 86 n n n n n31 XRT 184 9·3 n n y n n32 XRT 250 29·6 n n n n n33 XRT 249 24·8 n n n n n34 XRT 175 12 n n n n n35 uni adren 29 24·9 n n n n n36 uni adren 22 35·3 n n n n n37 uni adren 12 12·4 n n n n n

tsp, transsphenoidal pituitary surgery; XRT, Pituitary irradiation; uni adren, unilateral adrenalectomy; N/A, not assessed; y, yes; n, no.

Hormonal assays

The GH assays were performed by an in-house radioimmu-noassay until 1986 when the method was changed to a two-siteimmunoradiometric assay. Both assays used the same referencepreparation (NIBSC 66/217) and gave highly comparableresults. All assays performed after 1990 were measured againstthe reference preparation NIBSC 80/505. This produced resultsthat differed by a factor of 1·2 and so results were accordinglymodified to permit comparison of values across the whole timespan of the study.

Results

Surgically cured patients

Twenty patients (four male) with surgically induced remissionof Cushing’s disease underwent assessment of GH secretorystatus (Table 1).

Seventeen underwent assessment of GH status less thantwo years following remission, of whom 10 (59%) wereseverely GH deficient at the earliest assessment (mean 9·3,range 0–21 months). Nine patients underwent assessment ofGH status beyond two years following remission (mean 60,range 28–113 months). Of these, six had already been tested inthe first two years following remission and found to be GHdeficient. When these six were subsequently retested greaterthan two years following remission, only two remainedseverely GH deficient. Thus in four patients recovery ofadequate GH reserve was demonstrated. The remaining twopatients who had not undergone previous assessments of GHsecretory status both had adequate GH reserve. Thus 22% (2/9)of patients tested in the period greater than two years followingsurgical cure of Cushing’s disease were severely GH deficient.

However, the latter subject group is biased as patients inwhom GH deficiency has been documented at earliestassessment are over-represented. Clearly, patients in whomadequate GH reserve has been documented within the first twopostoperative years are not likely to be re-tested later as there isno evidence to suggest that there is any further progressiveimpairment of GH status with time after the initial surgicalinsult. To attempt to obtain useful information on the number ofpatients who will remain severely GH deficient followingsurgical cure of Cushing’s disease it is more accurate toestimate the incidence of persistent severe GH deficiencyfollowing pituitary surgery by incorporating the data frompatients in whom the original testing demonstrated adequateGH reserve. Collating such data, 13% (2/15) of patients hadpersistent severe GH deficiency following surgically inducedcure of Cushing’s disease.

The 20 patients within the surgically treated cohort under-went a total of 43 GH provocative tests, which on 35 occasions

was an ITT. Ten patients were subjected to a single GHprovocative test and 10 patients had multiple (range, 2–5) GHprovocative tests. Of the patients in whom more than one GHprovocative test was performed, six underwent an ITT on eachoccasion, whilst four patients had different GH provocativetests: in two, an AST followed an ITT (one patient demon-strated recovery of GH secretory status and one remainedseverely GH deficient), in one patient an ITT followed an AST(in this patient GH secretory status recovered), and the finalpatient underwent four GH provocative tests (two ITT, oneAST and lastly a GST) and failed to demonstrated any recoveryof GH secretory status.

Across all time periods, there were five patients in whomsevere GH deficiency resolved, the median time to recovery ofGH secretory status in these patients was 19 months followingremission.

In the cohort as a whole, seven (35%) patients had anteriorpituitary hormone deficits other than GH deficiency: 21·4%(3/14) of patients with normal GH secretory status at the lastassessment, 83% (5/6) of patients with severe GHD at the lastassessment. Of the five patients who demonstrated recoveryof GH secretory status 40% (two) had additional anteriorpituitary hormone deficits. 16 of 20 patients (80%) requiredno further treatment for Cushing’s disease, in the remainingfour patients bilateral adrenalectomy was required to controlthe disease.

Patients cured with pituitary radiotherapy

Fourteen patients (2 male) underwent GH provocative testingfollowing radiotherapy induced remission of Cushing’s disease.GH provocative tests (13 ITT, one GST) were performed amean of 99 (range 4–250) months following remission. Thirty-six percentage (5/14) of patients demonstrated severe GHdeficiency (table one).

Within the radiotherapy treated cohort 14% (2/14) of patientsdemonstrated additional anterior pituitary hormone deficits:11% (1/9) of patients with normal GH secretory status and 20%(1/5) of patients with severe GH deficiency.

Patients with adrenal adenomas

Three patients (all female) with adrenal adenomas cured byunilateral adrenalectomy underwent assessment of GH secre-tory status. The assessments were performed a mean of 21(range 12–29) months following remission. All patients hadnormal GH secretory status.

Discussion

These data reveal the incidence of severe GH deficiency

64 N. R. Hughes et al.

q 1999 Blackwell Science Ltd,Clinical Endocrinology, 51, 61–66

following pituitary radiotherapy to be 36% (5/14), assessed inthe period 4–250 months following irradiation. This is a lowincidence compared with previous studies (Littleyet al., 1990;Murayama et al., 1992) in which the incidence of GHdeficiency following radiotherapy for Cushing’s disease wasreported to be 50%. However, Littleyet al. (1990) did notdefine severe GH deficiency but simply employed a peak GHresponse to provocative testing of 20 mU/l borrowed from thepaediatric literature as the threshold for the diagnosis of GHdeficiency. The assessment of GH status employed byMurayama et al. (1992) was the GH response to GHRH,compared with normative data provided by Shibasakiet al.(1984). Although providing some information on GH secretorystatus, this does not provide detailed data on the number ofpatients with severe GH deficiency. In general however, othercentres (Murayamaet al., 1992) employ higher doses ofradiotherapy than that used in our study, which is likely to meanthat our results are an underestimate of the world-wideincidence of radiation-induced GH deficiency followingconventional radiotherapy for Cushing’s disease.

The incidence of severe GH deficiency in the first two yearsfollowing surgically induced remission of Cushing’s disease is59% (10/17). However, recovery of normal GH status is seenfrequently as only 22% (2/9) of patients demonstrated GHdeficiency beyond 24 months following remission. Indeed, thetrue incidence of severe persistent GH deficiency can beestimated as only 13% (2/15) when data from all time periodsare collated. Studies to date (Tyrrellet al., 1977; Kuwayamaet al., 1981; Sempleet al., 1984; Burkeet al., 1990; Magiakouet al., 1994) have reported that the incidence of GHinsufficiency varies between 25 and 100%. However, compar-ison between these studies and our own is hampered by the lessstrict definition of GH deficiency that they have employed (apeak GH response to provocative testing varying between up to20 or 30 mU/l).This is likely to be responsible, in part for thelower incidence of GH deficiency demonstrated in this presentstudy. As GH replacement therapy is at present restricted toadults with severe GH deficiency and it is these adults who havebeen proven to benefit from therapy, it is the number of patientswith severe GH deficiency that provides the most valuableinformation related to clinical management.

Hypercortisolaemia is known to affect the GH/IGF-I axis:Spontaneous GH release is reduced, as are GH responses toarginine, hypoglycaemia and exercise, IGF-I levels areunchanged but there is evidence of peripheral IGF-Iresistance (Magiakouet al., 1994) and cell proliferation atthe growth plate is reduced (Baronet al., 1994). There is alsoevidence that despite surgical cure of Cushing’s disease, theeffects of hypercortisolaemia on GH release continue for atleast a year (Magiakouet al., 1994). Perhaps of mostimportance with respect to clinical practice, this current study

demonstrates that there is substantial recovery in GHsecretory status when patients are reassessed after24 months following remission, substantially longer than hasbeen previously studied. This has implications for themanagement and investigation of the GH status of thesepatients in the postoperative period. Furthermore, it isdebatable whether there are any long-term consequences oftemporary GH suppression in the period immediatelyfollowing surgical cure of Cushing’s disease, in thatMagiakou et al. (1994) report normal growth in childrenduring the first postoperative year despite suppressed GHsecretion following surgical cure of Cushing’s disease.

It would be useful to be able to predict which surgicallytreated patients in whom GH deficiency is demonstrated at thefirst assessment will remain GH deficient and whom willdemonstrate recovery of GH secretory status. This wouldfacilitate investigation and treatment of those patients thatdevelop persistent GH deficiency. One might predict thatpatients with additional pituitary hormone deficits may be morelikely to develop persistent GH deficiency, and indeed this issuggested by our data: 83% (5/6) of patients with GH deficiencyat the last assessment had additional anterior pituitary hormonedeficits, but only 40% (2/5) of patients in whom recovery of GHsecretory status was demonstrated had additional pituitaryhormone deficits.

The results of this study raise issues concerning earlierstudies describing the effects of growth hormone replacementin GH deficient patients (Salomonet al., 1989). In this latterstudy, patients that were GH deficient following treatment forCushing’s disease constituted a large proportion of the studypopulation (50% of the patients in the GH treatment group and25% of patients in the control group). It is noteworthy that inthese subjects, GH status may have been defined as little as oneyear following remission. Thus, it is quite possible that theymay have been only temporarily GH deficient as a result ofcontinued hypercortisolaemic suppression rather than sufferingfrom persistent treatment-induced GH deficiency.

In summary, the incidence of severe persistent GH deficiencyfollowing surgically induced or radiotherapy induced cure ofCushing’s disease is lower than has been suggested by previousstudies, at 13% and 36%, respectively, although these latterstudies have assessed GH insufficiency and not severe GHdeficiency. Recovery of GH secretory status is seen in a highproportion of patients reassessed, particularly those with noadditional pituitary hormone deficits, at a median time of19 months following surgically induced remission of Cushing’sdisease. Thus, we recommend that definitive assessment of GHsecretory status is delayed for at least two years followingsurgical cure of Cushing’s disease. This has importantimplications for patients in whom GH replacement therapy isbeing considered.

GH status in treated Cushing’s disease 65

q 1999 Blackwell Science Ltd,Clinical Endocrinology, 51, 61–66

References

Baron, J., Klein, K.O., Collie, M.J., Yanovski, J.A., Novosad, J.A.,Bacher, J.D. & Cutler, G.B. Jr (1994) Catch-up growth afterglucocorticoid excess: a mechanism intrinsic to the growth plate.Endocrinology, 135,1367–1371.

Binnerts, A., Deurenberg, P., Swart, G.R., Wilson, J.H. & Lamberts,S.W. (1992) Body composition in growth hormone-deficient adults.American Journal of Clinical Nutrition, 55, 918–923.

Burke, C.W., Adams, C.B., Esiri, M.M., Morris, C. & Bevan, J.S.(1990) Transsphenoidal surgery for Cushing’s disease: does what isremoved determine the endocrine outcome?Clinical Endocrinology,33, 525–537.

Cuneo, R.C., Salomon, F., Wiles, C.M., Round, J.M., Jones, D., Hesp,R. & Sonksen, P.H. (1992) Histology of skeletal muscle in adultswith GH deficiency: comparison with normal muscle and response toGH treatment.Hormone Research, 37, 23–28.

Frantz, A.G. & Rabkin, M.T. (1964) Human growth hormone: clinicalmeasurement, response to hypoglycaemia and suppression bycorticosteriods.New England Journal of Medicine, 271, 1375–1381.

Growth Hormone Research Society. (1998) Consensus guidelines forthe diagnosis and treatment of adults with growth hormonedeficiency: summary statement of the Growth Hormone ResearchSociety Workshop on Adult Growth Hormone Deficiency.Journal ofClinical Endocrinology and Metabolism, 83, 379–381.

Jorgensen, J.O., Pedersen, S.A., Thuesen, L., Jorgensen, J., IngemannHansen, T., Skakkebaek, N.E. & Christiansen, J.S. (1989) Beneficialeffects of growth hormone treatment in GH-deficient adults.Lancet,1, 1221–1225.

Kuwayama, A., Kageyama, N., Nakane, T., Watanabe, M. & Kanie, N.(1981) Anterior pituitary function after transsphenoidal selectiveadenomectomy in patients with Cushing’s disease.Journal ofClinical Endocrinology and Metabolism, 53, 165–173.

Littley, M.D., Shalet, S.M., Beardwell, C.G., Ahmed, S.R. & Sutton,M.L. (1990) Long-term follow-up of low-dose external pituitary

irradiation for Cushing’s disease.Clinical Endocrinology, 33, 445–455.

Magiakou, M.A., Mastorakos, G., Gomez, M.T., Rose, S.R. &Chrousos, G.P. (1994) Suppressed spontaneous and stimulatedgrowth hormone secretion in patients with Cushing’s diseasebefore and after surgical cure.Journal of Clinical Endocrinologyand Metabolism, 78, 131–137.

McGauley, G.A. (1989) Quality of life assessment before and aftergrowth hormone treatment in adults with growth hormone deficiency.Acta Paediatrica Scandinavica, 356 (Suppl.), 70–72.

Murayama, M., Yasuda, K., Minamori, Y., Mercado Asis, L.B.,Yamakita, N. & Miura, K. (1992) Long term follow-up of Cushing’sdisease treated with reserpine and pituitary irradiation.Journal ofClinical Endocrinology and Metabolism, 75, 935–942.

Salomon, F., Cuneo, R.C., Hesp, R. & Sonksen, P.H. (1989) The effectsof treatment with recombinant human growth hormone on bodycomposition and metabolism in adults with growth hormonedeficiency.New England Journal of Medicine, 321,1797–1803.

Semple, C.G., Thomson, J.A. & Teasdale, G.M. (1984) Transsphenoi-dal microsurgery for Cushing’s disease.Clinical Endocrinology, 21,621–629.

Shibasaki, T., Shizume, K., Nakahara, M., Masuda, A., Jibiki, K.,Demura, H., Wakabayashi, I. & Ling, N. (1984) Age-related changesin plasma growth hormone response to growth hormone-releasingfactor in man.Journal of Clinical Endocrinology and Metabolism,58, 212–214.

Thorner, M.O., Bengtsson, B.A., Ho, K.Y., Albertsson Wikland, K.,Christiansen, J.S., Faglia, G., Irie, M., Isaksson, O., Jorgensen, J.O.,Ranke, M., Rosenfield, R., Shalet, S., Takano, K., van der Veen, E.A.& Weissberger, A. (1995) The diagnosis of growth hormonedeficiency (GHD) in adults [letter].Journal of Clinical Endocrinol-ogy and Metabolism, 80, 3097–3098.

Tyrrell, J.B., Wiener Kronish, J., Lorenzi, M., Brooks, R.M. &Forsham, P.H. (1977) Cushing’s disease: growth hormone responseto hypoglycemia after correction of hypercortisolism.Journal ofClinical Endocrinology and Metabolism, 44, 218–221.

66 N. R. Hughes et al.

q 1999 Blackwell Science Ltd,Clinical Endocrinology, 51, 61–66