Acromegalia 3

7/27/2019 Acromegalia 3

1/6

Significance of Abnormal Nadir Growth HormoneLevels after Oral Glucose in Postoperative Patients with Acromegaly in Remission with Normal Insulin-LikeGrowth Factor-I Levels

PAMELA U. FREDA, ABU T. NURUZZAMAN, CARLOS M. REYES, ROBERT E. SUNDEEN, ANDKALMON D. POST

Department of Medicine, Columbia University College of Physicians and Surgeons, New York, New York 10032;and Department of Neurosurgery, Mount Sinai Medical Center, New York, New York 10029

Our initial study in postoperative patients with acromegalyidentified a group of patients in remission, as defined by nor-mal IGF-I levels, but who had a subtle abnormality of GHsuppression after oral glucose. To investigate the significanceof this abnormality, we haveundertaken further detailed test-ing of GH secretion and a longitudinal follow-up of some of these patients.

Of the 110 postoperative patients with acromegaly evalu-ated by oral glucose tolerance test, 76 were in remission ( i.e.normal IGF-I level), and of these subjects with acromegaly inremission, 50 had normal nadir GH ( < 0.14 g/ml) (group I),and 26 had abnormal nadir GH ( > 0.14 g/ml) (group II). Four-teen subjects in remission, seven from remission group I andseven from remission group II, underwent additional testing consisting of both hourly GH sampling over 8 h and, on aseparate day, arginine stimulation testing. The mean of hourly GH was higher in group II (0.47 0.04 g/liter) than ingroup I (0.19 0.07 g/liter; P 0.002). GH response to arginine

was greater in group II than in group I ( P < 0.01). Of thosepatients in remission from the initial cohort studied, 49 (30subjects from group I and 19 from group II) underwent seriallongitudinal oral glucose tolerance testing every 12 yr overa 1- to 6.5-yrperiod (mean follow-up, 3.2yr). Theinitialpatternof GH suppression persisted in most patients. IGF-I levelsremained normal in all patients in group II, but five subjectsfrom group II developed an elevated IGF-I level and, thus, abiochemical recurrence. The rate of disease recurrence wasgreater in group II than in group I ( P 0.003).

We have found that some postoperative subjects with ac-romegaly in remission with normal IGF-I levels have persis-tentlyabnormalnadirGH levelsafteroral glucose that maybeaccompanied by other evidence of greater GH secretion thanpostoperative patients with normal GH suppression. This ab-normal pattern of GH suppression may be associated withincreased risk of disease recurrence in some patients. ( J Clin Endocrinol Metab 89: 495500, 2004)

BIOCHEMICAL CRITERIAFORremission of acromegalyhave become more stringent with the development of highly sensitive and specific GH assays and their use inconjunction with measurement of serum IGF-I levels. Use of these assays has revealed that nadir GH levels after oralglucose in patients with active acromegaly with elevatedIGF-I levels may be much lower than previously could beappreciated with less sensitive assays and, in some cases,may be less than 1 g/liter (1, 2). Polyclonal RIAs for GHwere also insufficiently sensitive to discriminate GH levelsaround the cut-off of 2 g/liter that was in use with theseassays, but with modern assays, discrimination of oral glu-cose tolerance testing (OGTT) nadir GH levels in healthysubjects from those of patients with active acromegaly isimproved (1). However, even with theuse ofmodern GHandIGF-I assays, the results of these two tests are not alwayscongruent. In a prior study, we evaluated serum IGF-I levelsin conjunction with oral glucose-suppressed GH levels witha highly sensitive and specific immunoradiometric assay

(IRMA) in postoperative patients in comparison witha groupofhealthy subjects (1). In this study,we identified a subgroupof postoperative patients in remission who had normal se-rum IGF-I levels yet whose GH did not suppress completelyinto the range of the healthy subjects comparison group. Theclinicalsignificanceof abnormalsuppression in suchpatientswas uncertain, but we hypothesized that it could representGH secretory dysregulation and/or mild excess GH secre-tion. It was also uncertain whether this pattern of subtle GHabnormality could be a precursor to clinically significantrecurrent active acromegaly. To further investigate thesepossibilities, we undertook further detailed investigation of GH secretion and longitudinal follow-up of some membersof the initial cohort of postoperative patients in remission.

Subjects and Methods Study subjects

Initial acromegaly cohort.A total of 110 patients who had undergonetranssphenoidal surgery for GH-secreting pituitary tumors were eval-uated at least 6 monthsaftersurgery.The cohortconsisted of 60 menand50 women, mean age was 46 yr (range, 1777 yr), and mean body massindex (BMI) was 26.7 kg/m 2 . All were outpatients and did not havehepatic or renal disease, glucose intolerance, or diabetes mellitus. Base-line data on the first 60 (1) and then 95 subjects (3) were previouslyreported. Subjects were categorized into those with active disease (el-evated IGF-I) and those in remission (normal IGF-I), and then remission

Abbreviations: BMI, Body mass index; IRMA, immunoradiometricassay; OGTT, oral glucose tolerance test. JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the en-docrine community.

0021-972X/04/$15.00/0 The Journal of Clinical Endocrinology & Metabolism 89(2):495500 Printed in U.S.A. Copyright 2004 by The Endocrine Society

doi: 10.1210/jc.2003-031316

495


2/6

group I (normal GH suppression) and remission group II (abnormal GHsuppression) (Table 1).

Hourly GH sampling and arginine testing subjects.All subjects who werefound to be in remission from the initial cohort (normal IGF-I levels)were invited to participate in these studies. Those who had prior pitu-itary radiotherapy were excluded. Fourteen subjects agreed to partici-pate: seven from remission group I (normal GH suppression) and seven

from remission group II (abnormal GH suppression). Of the postmeno-pausal women studied, subject 3 was receiving Premarin, and subjects9 and 10 were not on hormone replacement therapy. All other femaleswere studied in the early follicular phase of their menstrual cycles.Characteristics of the study subjects are shown in Table 2.

Longitudinal follow-up cohort.All patients from the initial cohort whowere in remission (n 76) were also invited to come for longitudinalevaluations and testing. Only the patients willingness to participatedetermined the study group. Forty-nine patients agreedto participate inthis follow-up, 30 patients from remission group I (normal GH sup-pression) and 19 from remission group II (abnormal GH suppression).Group I consisted of 16 women and 14 men [mean age 46.7 2.2 yr;range, 25 77 yr] whose first evaluation was conducted at a mean of 2.58 .53 yr after surgery. Group II consisted of 12 women and 7 men[mean age, 48 3.2 yr; range, 19 75 yr] who were first evaluated at amean of 4.8 0.82 yr after surgery. No patients in group I or II werereceiving medical therapy for acromegaly during the follow-up period.Mean BMIof subjects was28.1kg/m 2 ingroupIand27.2kg/m 2 in groupII (P 0.40).

Healthy subjects.Forty-sixhealthy subjects (26men, 20 women;meanage,40 yr; range, 19 71 yr) were also studied. Mean BMI was 23.4 kg/m 2 .Data on this group were previously reported (3).

Premenopausal healthy women and women with acromegaly werestudied in the early follicular phase of the menstrual cycle (d 1 5).

Study procedures

Procedures are described for each of the study groups above.

OGTT. OGTT was performed after an overnight fast. While seated,subjects had blood sampled at baseline (fasting) and then at 60, 90, and120 min after drinking 100 g dextrose (Trutol 100 Glucose ToleranceBeverage, NERL Diagnostics, East Providence, RI). Blood was allowedto clot at room temperature for 15 min; it was then centrifuged, and theserum was frozen at 80 C in multiple aliquots. Fasting blood sampleswere assayed for IGF-I. Blood samples at all time points were assayedfor GH levels. Serum glucose levels at baseline and at 2 h post dextrosewere less than 120 mg/dl and less than 200 mg/dl, respectively, in allsubjects.

Hourly GH sampling and arginine testing.Subjects underwent the follow-ing tests in random order at least 1 wk apart.

Hourly GH sampling. Patients were admitted to our clinical researchcenter at 0800 h after an overnight fast, and blood was sampled throughan indwelling forearm catheter hourly from 0900 1700 h. Subjects werefed breakfast at 0900 h and lunch at 1200 h. Meals were standardized to29% fat, 56% carbohydrate, and 15% protein. Subjects remained seatedor recumbent during the testing.

Arginine stimulation testing. Subjects were admitted to our clinicalresearch center after an overnight fast and remained fasting during thetesting. Beginning at 0900 h, an iv infusion of 30 g l -arginine hydro-chloride (R-Gene 10, PharmaciaCorporation, Kalamazoo,MI) was givenover 30 min. Blood was sampled before and at 30, 60, 90, and 120 minafter the infusion.

Longitudinal follow-up study of patients in remission.Patients were evalu-

ated serially every 1 2 yr over a 1- to 6.5-yr period with laboratoryevaluations consisting of an OGTT for GH and IGF-I levels as describedabove.

Healthy subjects.Healthy subjects underwent OGTT once as describedabove.

These protocols were approved by the Institutional Review Board of Columbia-Presbyterian Medical Center, Columbia University (NewYork, NY), and informed consent was obtained from all subjects.

Assays

GH. GH was measured by a two-site IRMA obtained from DiagnosticSystems Laboratories (Webster, TX). The standards for this IRMA con-tain 22K recombinant human GH andare calibrated to the World HealthOrganization (WHO) International Reference Preparation of humanGH(code 88/624), the most recent calibrator from the WHO. There is nocross-reactivity with otherhuman pituitary hormones,includinghumanprolactin or with other species of GH. The intraassay coefficient of variation is 3.1%, and the interassay coefficient of variation is 5.9%. Theassay sensitivity in our laboratory is 0.05 g/liter. The upper limit of normal for nadir GH level after oral glucose in our laboratory is 0.14

g/liter (1, 3).

IGF-I. IGF-I was measured by RIA using a polyclonal rabbit antibodygenerated against human IGF-I obtained from Nichols Institute (San Juan Capistrano, CA). In this assay, soluble IGF-I is separated from its binding proteins by extraction with acid-ethanol and precipitated at 20C. Recombinant human IGF-I is used for the standards and labeled withI125 for the tracer. The antiserum for IGF-I shows virtually no cross-reactivity with IGF-IIor GH.The standard is calibrated against WHO1stInternational Reference Reagent 1988, IGF-I 87/518. The intraassay co-efficient of variation is 4%, and the interassay coefficient of variation is

11%. Assay sensitivity is 13.5 g/liter. The normal ranges for this assayare as follows: age 16 24 yr, 182780 g/liter; 25 39 yr, 114 492 g/liter; 40 54yr,90 360 g/liter; above 55 yr,71 290 g/liter.All subjects IGF-I levels were compared with their age-appropriate normal ranges.

Serum glucose was measured by the hexokinase method.

Statistical analysis

Nadir GH was defined as the lowest value at any time point after oralglucose administration.The upperlimitof normalnadirGH wasdefinedastheGH value 2 sd valuesabove the mean nadir of the healthy subjectsas measured by IRMA (0.14 g/liter) (1, 3). Baseline and peak GH levelsafter arginine stimulation were compared within each group by pairedStudent s t test. GH responses to arginine testing in remission group Ivs. group II were compared by repeated measures ANOVA. The meansof hourly GH levels were calculated for each subject, and those forgroups I and II were compared by ANOVA. Follow-up periods of

TABLE 1. Mean nadir GH ( SE ) after glucose and mean IGF-I levels in healthy subjects and subjects with acromegaly

Healthy subjects(n 46)

Acromegaly active(n 34)

Acromegaly remissiongroup I (n 50)

Acromegaly remissiongroup II (n 26)

GH levels (IRMA) ( g/liter)Mean nadir 0.08 0.01 1.34 0.23 0.10 0.01 0.30 0.05Range 0.05 0.13 0.33 5.0 0.05 0.14 0.15 1.2

IGF-I levels ( g/liter)Mean 240 13 716 50 255 14 293 18Range 106 449 328 1356 120 679 138 533

Subjects with acromegaly were first divided based on IGF-I level into active disease (elevated IGF-I) and remission (normal IGF-I). Subjectsin remission were also divided into remission group I (normal GH suppression, i.e. nadir GH 0.14 g/liter) and remission group II (abnormalGH suppression, i.e. nadir GH 0.14 g/liter).

496 J Clin Endocrinol Metab, February 2004, 89(2):495 500 Freda et al . OGTT in Acromegaly


3/6

groups I and II were compared by ANOVA. The recurrence rate ingroups I and II were compared by Fisher s exact test. GH levels arereported as mean se .

Results Baseline evaluation of the cohort and definition of remission groups

As shown in Table 1 and as we have previously reported(1, 3), postoperative subjects with acromegaly were first di-vided based on IGF-I level into those in remission (normalIGF-I) and those with active disease (elevated IGF-I) level.Subjects in remission were also divided into group I ( nor-mal GH suppression, i.e. nadir GH 0.14 g/liter) andgroup II ( abnormal GH suppression, i.e. nadir GH 0.14

g/liter) (Fig. 1). IGF-I levels in groups I and II were exam-ined in relation to quartiles within the normal IGF-I range.Quartiles for mean IGF-I levels in groups I and II were notdifferent.

Hourly GH sampling and arginine stimulation testing of

remission subjects

Arginine stimulation. Responses to arginine stimulation ingroup I vs. group II are shown in Fig. 2 and Table 2. Meanof baseline GH in group I was 0.19 0.06 g/liter and roseto a mean peak of 1.1 0.32 g/liter ( P 0.04). Mean GHat baseline in group II was 0.45 0.10 g/liter and rose toa mean of 2.78 0.88 g/liter ( P 0.04). Mean GH duringGH arginine testing GH to arginine testing was significantlyhigher in group II (1.47 0.28 g/liter) vs. group I (0.470.11 g/liter) ( P 0.01).

Hourly GH sampling

The means of hourly GH measurements for the subjects ingroup I vs. group II are shown in Fig. 2 and Table 2. Themeans of hourly GH measurements over the day were sig-nificantly higher in group II (0.47 0.04 g/liter) vs. groupI (0.19 0.07 g/liter) ( P 0.002).

T A B L E 2

. C l i n i c a l a n d l a b o r a t o r y c h a r a c t e r i s t i c s o f t h e s u b j e c t s w i t h a c r o m e g a l y i n r e m i s s i o n w h o u n d e r w e n t h o u r l y G H s a m p l i n g a n d a r g i n i n e t e s t i n g

P a t i e n t

n o .

A g e

( y r )

S e x

B M I

( k g / m

2 )

P r e o p e r a t i v e

t u m o r s i z e

Y e a r s

s i n c e

s u r g e r y

R a d i o t h e r a p y

H y p o p i t u i t a r i s m

O t h e r

e n d o c r i n e

f u n c t i o n

N a d i r G H

d u r i n g

O G T T

( g / l i t e r )

I G F - I

( n g / m l )

M e a n G H

o n h o u r l y

s a m p l i n g

( g / l i t e r )

B a s a l G H

a r g i n i n e

t e s t i n g

( g / l i t e r )

P e a k G H

a r g i n i n e

t e s t i n g

( g / l i t e r )

G r o u p I

1

3 8

F

2 5

M a c r o

9

N o

N o

0 . 0 8

1 5 9

0 . 1 4

0 . 2 9

1 . 2

2

5 2

F

2 0 . 4

M i c r o

2 . 5

N o

N o

1 H y p o t h .

0 . 0 9

2 0 4

0 . 1 4

0 . 4 4

1 . 0

3

5 0

F

3 3

M a c r o

4

N o

N o

1 H y p o t h .

0 . 0 5

1 7 3

0 . 0 8

0 . 0 7

0 . 6

4

5 4

M

3 4

M a c r o

1

N o

N o

0 . 0 5

1 6 7

0 . 1 3

0 . 0 6

0 . 4

5

3 1

M

2 9 . 3

M a c r o

2

N o

N o

0 . 0 9

1 5 7

0 . 4 5

0 . 0 5

2 . 8

6

4 6

M

3 0 . 7

M a c r o

1 . 5

N o

N o

0 . 0 5

1 5 4

0 . 3 1

0 . 9 6

1 . 8

7

3 8

F

4 2

M a c r o

7

N o

N o

0 . 0 8

2 0 8

0 . 0 8

0 . 2 7

0 . 2 8

G r o u p I I

8

3 9

M

3 1 . 5

M a c r o

2

N o

2 H y p o g o n .

0 . 3 2

3 1 8

0 . 4 3

0 . 3 7

1 . 0

9

5 4

F

2 3

M i c r o

1 0

N o

N o

0 . 2 3

1 3 1

0 . 5 2

1 0

5 7

F

2 2 . 9

M a c r o

1 0

N o

N o

1 H y p o t h .

0 . 3 0

2 9 0

0 . 4 1

0 . 2 5

1 . 3

1 1

5 0

M

2 4 . 3

M a c r o

1 . 5

N o

2 H y p o g o n .

0 . 2 4

2 9 2

0 . 6 6

0 . 4 9

7 . 4

1 2

6 6

M

2 5 . 8

M a c r o

7

N o

N o

0 . 4 3

1 4 2

0 . 4 3

0 . 1 9

4 . 0

1 3

7 9

M

2 5

M a c r o

1 1

N o

N o

0 . 4 0

1 2 2

0 . 5 6

3 . 0

1 4

2 6

F

3 0 . 9

M a c r o

2 . 5

N o

N o

0 . 2 3

1 6 8

0 . 3 6

0 . 9 6

1 . 4

1 H y p o t h .

, P r i m a r y h y p o t h y r o i d i s m ; 2

H y p o g o n . ,

s e c o n d a r y h y p o g o n a d i s m

.

F IG . 1. Nadir GH levels after oral glucose and IGF-I levels in post-operative patients in remission as defined by normal IGF-I levels. ,Remission group I, nadir GH levels within the range of healthy subjects ( 0.14 g/liter). F , Remission group II, nadir GH levels wereabove the normal range ( 0.14 g/liter).

Freda et al . OGTT in Acromegaly J Clin Endocrinol Metab, February 2004, 89(2):495 500 497


4/6

Longitudinal follow-up study of patients in remission

Patients in group I were followed longitudinally over amean of 2.9 0.32 yr (range, 1 6.5 yr), and those in groupII for a mean of 3.7 0.36 yr (range, 1 5 yr). The follow-up

periods of the two groups were not significantly different(P 0.14). The time from surgery to the initial OGTT eval-uation was greater in group II (4.8 0.82 yr) vs. group I(2.58 0.53 yr) (P 0.02), possibly suggesting that theabnormal pattern of GH suppression develops with timeafter surgery. However, this cannot be determined fromthese data, because not all of these patients were followedstarting immediately after surgery. In group II, 7 of 19 pa-tients had undergone previous radiotherapy, vs. 2 of 30 ingroup I; a greater proportion of subjects in group II had priorradiotherapy ( P 0.019).

Serial OGTT testing revealed that the pattern of GH sup-pression found on initial testing persisted on follow-up inmost cases. Of the patients in group I, 29 of 30 continued to

have normal GH suppression, and 1 of 30 developed abnor-mal GH suppression (nadir of 0.14 g/liter to 0.19 g/liter).IGF-I levels remained normal over the follow-up in all pa-tients in group I.

Of the patients in group II, nadir GH was persistentlyabnormal in 17 and normalized in two. Regarding the twopatients in group II whose nadir normalized: the first pa-tient s nadir fell from 0.18 g/liter to 0.07 g/liter, and IGF-Ifell from 300 ng/ml to 132 ng/ml over a 3-yr period; thesecond patient s nadir GH fell from 0.15 g/liter to 0.09

g/liter, and IGF-I fell from 318 ng/ml to 102 ng/ml over a6-yrfollow-up period.Bothpatientshad undergone previousradiotherapy.

IGF-I levels remained normal over the follow-up period in14 subjects in group II. However, five subjects from group IIdeveloped an IGF-I level of 15% above their age-adjustedupper limit of normal along with continued abnormal GHsuppression. New, persistent elevation of IGF-I level wasconsidered to represent a biochemical recurrence.The rate of disease recurrence in group II was significantly higher thanin group I ( P 0.003). Recurrence was detected 2 6 yr aftersurgical therapy. None of the five patients who developed arecurrence had received prior radiotherapy. IGF-I levels dur-ing the follow-up period before recurrence were in the upperhalf of the normal range in all five patients and in the upperquartile in three of five patients. In two of five patients,visible tumor regrowth was demonstrated on magnetic res-onance imaging at the time of biochemical recurrence.

Discussion

Recent data have shown that GH levels measured withhighly sensitive and specific GH assays should fall to less

than 1 g/liter in most normal subjects after an oral glucoseload (1, 4). In a group of healthy subjects that we studied,nadir GH levels were all below 0.14 g/liter (3). In compar-ison to these healthy subjects, of the cohort we studied, mostpostoperative patients in remission with normal IGF-I levelshad nadir GH levels within the healthy subjects range, buta subset of postoperative patients in remission did not. Ad-ditional evaluation of these patients in remission has shownthat the pattern of GH suppression present on initial testingpersists over time in most patients, thus validating this as areproducible marker of the GH axis. Additionally, we foundthat those patients with abnormal GH suppression also hadhigher mean GH levels and greater GH stimulation after

arginine than those in remissionwith normalnadir GH. Withlongitudinal follow-up, abnormal GH nadir was associatedwith a higher rate of disease recurrence as defined by thedevelopment of elevation of IGF-I level.

Failure of normal GH suppression after oral glucose is awell-known characteristic of patients with active acromeg-aly. These patients may have no change, a partial suppres-sion, or a paradoxical increase in GH in response to oralglucose administration (5, 6). The mechanism for GH sup-pression in healthy subjects is not clear, but it may be due inpart to somatostatin release in response to the oral glucoseload. In acromegaly, the etiology of abnormal GH suppres-sion is also not clear but may be an impaired somatostatinresponse (7) or some tumoral resistance to suppression by

F IG . 2. Top, Mean of hourly GH measurements in patients in remis-sion groups I ( ) and II ( F ). Bottom, Mean of GH response to argininetesting at each time point measured in patients with acromegaly inremission groups I and II.



5/6

somatostatin. The subtle impairment of GH suppression weobserved in remission group II was reproducible over time,suggesting that this represents persistent GH dysregulationin these patients. Most patients in group II (20 of 26) also hadnadir GH levels that were belowthose of patients with activeacromegaly (high IGF-I), consistent with an intermediarydegree of GH dysregulation in group II. Also, a larger per-centage of patients in group II than in group I had undergoneprevious radiotherapy, consistent also with this pattern of GH suppression, representing a transition from active tonormal GH secretion. We did consider and exclude otherpotential causes of abnormal GH suppression (other thanactiveacromegaly) in our cohort such as glucose intolerance,diabetes mellitus, chronic renal insufficiency, liver failure,active hepatitis, hyperthyroidism, anorexia nervosa, andother forms of malnutrition (8). Obesity has also been re-ported to be associated with less glucose suppression of GHlevels than in lean subjects (9, 10), but BMIs were not sig-nificantly different in remission groups I and II. All subjectswere evaluatedat least6 monthspostoperatively,when earlypostoperative changes in GH or IGF-I levels should not have been a factor (11).

In addition to the subtle impairment in GH regulation viathe glucose-suppression pathway, we have also found thatpatients in remission group II have other evidence of mildGH excess relative to those postoperative patients in remis-sion group I. Group II patients have greater mean of hourlyGH levels and secrete more GH after arginine administrationthan those subjects in group I. Mean GH assessments have been used widely to assess disease status in patients withacromegaly. We cannot establish disease status based onthese mean GH measurements in our patients because, withmodern assays, mean GH values overlap in healthy and

acromegaly subjects, especially in patients with mild disease(2). However, the results of this study do indicate that pa-tients with abnormal GH suppression also secrete relativelymore GH over the day than those with normal GHsuppression.

It is well known that arginine administration may stim-ulate GH secretion, both in patients with GH-secreting ad-enomas (12) and healthy subjects. However, it is also knownthat the GH response to arginine infusion is not uniform inpatients with acromegaly (6). GH stimulation by argininemay occur via inhibition of somatostatin (13, 14), and someinvestigators have used arginine testing to assess the integ-rity of the somatostatin pathway. For example, the GH re-

sponse to arginine may be diminished in patients with ac-romegaly with prior radiotherapy, regardless of whether ornot they have other evidence of GH deficiency, implyingimpaired somatostatin tone due to radiotherapy (15, 16). Inour patients, none of those evaluated with arginine testinghad prior radiotherapy, eliminating this as a possible factorto the groups differential responses to arginine. In addition,none of the subjects in group I had any evidence of hypo-pituitarism. The purpose of this evaluation was not to eval-uate GH reserve, but rather to provide other evidence forrelatively greater GH secretion in groups I and II, which theresults do support.

In this study,using IGF-Ias themarkerfor disease activity,we found that some of those patients with abnormal GH

suppression developed over time a persistent elevation of IGF-Iand, thus, a biochemical recurrence at a higherrate thanthose patients with normal GH suppression. These data sug-gest that GH dysregulation has progressed to true GH excessas reflected in the elevation of IGF-I level in these patients.The serum IGF-I level is an excellent marker of overall GHsecretion anda very sensitive indicatorof GHexcess(8). Datain newly diagnosed patients have shown that IGF-Ielevationcan detect GH excess at nadir GH levels less than 1 g/liter,levels not previously thought to be compatible with acro-megaly (2).

IGF-I elevation also seems to be more specific than mean24-h assessments at detecting acromegaly, especially in pa-tients with mild degrees of GH excess, because mean 24-hGH levels in patients with acromegaly (with elevated IGF-Ilevels) can overlap with those of healthy subjects (2, 17, 18).Mean GH levels are not necessarily predictive of integratedperipheral GH effect, as reflected in the IGF-I level, becausethe pattern of GH secretion is also a determinant of IGF-Iproduction. Thus, despite similar mean GH levels, patientswith acromegaly, who have elevated trough or valley GHconcentrations, producehigher IGF-I levels thanhealthy subjects who have normal pulsatile GH secretion (17, 19). Witha few exceptions, including conditions that can lower IGF-Ilevel, such as malnutrition and liver disease, or raise IGF-I,such as pregnancy or adolescence, IGF-I is a very accuratepredictor of disease status in acromegaly (8).

We did find, however, that in remission group II, abnor-mal GH suppression persists in most patients despite a nor-mal IGF-I. These findings suggest that IGF-I alone may notreveal subtle degrees of GH dysregulation. The abnormalpattern of pulsatile GH secretion typical of acromegaly may be corrected by successful surgery (18). In other patients,

abnormalities suggestive of disordered GH neuroregulationcan persist despite a normal IGF-I level (17, 20), and suchpatients might be at increased risk for the return of activedisease with elevation of IGF-I levels (20). Also, because of the wide range of normal IGF-I levels, we cannot excludethe possibility that an IGF-I within the normal referencerange may not be truly normal for some individuals. Thispossibility further supports the value of examining GH sup-pression in addition to IGF-I levels. Other evidence of per-sistent GH dysregulation based on dynamic testing, such asparadoxical responses to TRH stimulation despite other bio-chemical evidence consistent with remission, may also bepredictive of postoperative recurrence in some patients (21,

22). Variability in GH response to TRH makes this test prob-lematic for routine use in postoperative assessment.Many studies have assessed recurrence rates after trans-

sphenoidal surgery for acromegaly, but these rates vary con-siderably, possibly because of lack of uniformity in the cri-teria used to define remission. It is likely that older studiesoverestimated the recurrence rate by including patients whoactually had persistent postoperative disease not detected byless precise GH assays. In a recent series, the rate of recur-rence was reported to be between 1.1 and 19% (23 27). Byidentification of characteristics of postoperative patients atgreatest risk for recurrence, these patients could be targetedwith more intense follow-up assessments. Our data are sug-gestive that GH suppression assessed with highly sensitive

Freda et al . OGTT in Acromegaly J Clin Endocrinol Metab, February 2004, 89(2):495 500 499


6/6

andspecific assayscould be such a characteristic.In addition,the fact that 5 of 12 of those postoperative patients withabnormal suppression who had not had radiotherapy wenton to a recurrence is further evidence of the significance of these findings. However, the follow-up period for some of the patients is still relatively short, so long-term follow-upof our remission cohorts should be undertaken to establishwhether thepattern ofGHsuppression canbe used to predictlong-term remission or an increased risk of recurrence aftersurgery for acromegaly.

In conclusion, patients with acromegaly in remission post-operatively whohavea subtleyet persistent failure ofnormalGH suppression have additional evidence of relatively in-creased GH secretion compared with those patients withnormal GH suppression. As long as IGF-I normalization ismaintained, these patients can be considered to be in remis-sion and can be observed without additional therapy. How-ever, with longitudinal follow-up, abnormal GH nadir wasassociated with a higher rate of disease recurrence as defined by the development of elevation in IGF-I level. Thus, in asmall number of postoperative patients, subtle GH neuro-regulation may persist despite clinical remission, and, if identified, these patients may need to be monitored moreclosely for recurrence.

Acknowledgments

We thank Ms. Anne Knieriem and the nursing staff and bionutritionunit of the Columbia University General Clinical Research Center(GCRC) for their assistance with these studies.

Received July 29, 2003. Accepted September 15, 2003.Address all correspondence and requests for reprints to: Pamela U.

Freda, M.D., Department of Medicine, Columbia University College of Physicians and Surgeons, 630 West 168th Street, New York, New York10032. E-mail: [email protected] work wassupportedby National Institutes of Health Grants K08DK02561 and R03 DK60475 (to P.U.F.) and RR00645 to the ColumbiaUniversity GCRC, and by Novartis International AG.

Results of this work were presented in part at the 84th Annual Meet-ing of The Endocrine Society, San Francisco, CA, 2002.

References

1. Freda PU, Post KD, Powell JS, Wardlaw SL 1998 Evaluation of disease statuswith sensitive measures of growth hormone secretion in 60 postoperativepatients with acromegaly. J Clin Endocrinol Metab 83:3808 3816

2. Dimaraki EV, Jaffe CA, DeMott-Friberg R, Chandler WF, Barkan AL 2002Acromegaly with apparently normal GH secretion: implications for diagnosisand follow-up. J Clin Endocrinol Metab 87:3537 3542

3. Freda PU, Landman RE, Sundeen RE, Post KD 2001 Gender and age in the biochemical assessment of cure of acromegaly. Pituitary 4:163 171

4. Chapman IM,Hartman ML,Straume M, Johnson ML,VeldhuisJD, ThornerMO 1994 Enhanced sensitivity growth hormone (GH) chemiluminescenceassay reveals lower postglucose nadir GH concentrations in menthan women. J Clin Endocrinol Metab 78:1312 1319

5. Earll JM,Sparks LL, Forsham PH 1967 Glucose suppression of serum growthhormone in the diagnosis of acromegaly. JAMA 201:628 630

6. Lawrence AM, Goldfine ID, Kirsteins L 1970 Growth hormone dynamics inacromegaly. J Clin Endocrinol Metab 31:239 247

7. Shibasaki T, Masuda A, Hotta M, Yamauchi N, Hizuka N, Takano K, De-

mura H, Shizume K 1989 Effects of ingestion of glucose on GH and TSHsecretion: evidence for stimulation of somatostatin release from the hypothal-amus by acute hyperglycemia in normal man and its impairment in acrome-galic patients. Life Sci 44:431 438

8. Freda PU 2003 Current concepts in the biochemical assessment of the patientwith acromegaly. Growth Horm IGF Res 13:171 184

9. Maccario M, Procopio M, Grottoli S, Oleandri SE, Razzore P, Camanni F,Ghigo E 1995 In obesity the somatotrope response to either growth hormone-releasing hormone or arginine is inhibited by somatostatin or pirenzepine but

not by glucose. J Clin Endocrinol Metab 80:3774 377810. Bonora E, Moghetti P, Zenere M, Querena M, Tosi F, Corgnati A, MuggeoM 1990 Plasma concentrations of growth hormone during hyperglycemicclamp with or without somatostatin infusion in obese subjects. J Clin Endo-crinol Metab 70:1732 1734

11. Kristof RA, Neuloh G, Redel L, Klingmuller D, Schramm J 2002 Reliabilityof the oral glucose tolerance test in the early postoperative assessment of acromegaly remission. J Neurosurg 97:1282 1286

12. Hanew K, SasakiA, Mouri T, YoshinagaK 1980 Plasmagrowth hormone andprolactin responses to TRH and arginine given before and during bromocrip-tine therapy to patients with acromegaly. Acta Endocrinol 95:298 307

13. Alba-Roth J, Muller OA,Schopohl J, von Werder K 1988 Arginine stimulatesgrowth hormone secretionby suppressing endogenous somatostatin secretion. J Clin Endocrinol Metab 67:1186 1189

14. Masuda A, Shibasaki T, Nakahara M, Imaki T, Kiyosawa Y, Jibiki K, De-mura H, Shizume K, Ling N 1985 The effect of glucose on growth hormone(GH)-releasing hormone-mediated GH secretion in man. J Clin EndocrinolMetab 60:523526

15. Peacey SR, Toogood AA, Shalet SM 1998 Hypothalamic dysfunction incured acromegalyis treatmentmodalitydependent. J ClinEndocrinol Metab83:16821686

16. MurrayRD, PeaceySR, RahimA, Toogood AA,ThornerMO,ShaletSM 2001The diagnosis of growth hormone deficiency (GHD) in successfully treatedacromegalic patients. Clin Endocrinol (Oxf) 54:37 44

17. Ho KY, Weissberger AJ 1994 Characterization of 24-hour growth hormonesecretion in acromegaly: implications for diagnosis and therapy. Clin Endo-crinol (Oxf) 41:7583

18. Hartman ML, Veldhuis JD, Vance ML, Faria AC, Furlanetto RW, ThornerMO 1990Somatotropinpulse frequencyand basalconcentrationsare increasedin acromegaly andare reduced by successful therapy. J Clin Endocrinol Metab70:13751384

19. Peacey SR, Toogood AA, Veldhuis JD, Thorner MO, Shalet SM 2001 Therelationship between 24-hour growth hormone secretion and insulin-likegrowth factor I in patients with successfully treated acromegaly: impact of surgery or radiotherapy. J Clin Endocrinol Metab 86:259 266

20. Ho PJ, Jaffe CA, Friberg RD, Chandler WF, Barkan AL 1994 Persistence of

rapid growth hormone (GH) pulsatility after successful removal of GH-producing pituitary tumors. J Clin Endocrinol Metab 78:1403 141021. ArafahBM, Rosenzweig JL,Fenstermaker R, Salazar R, McBride CE,Selman

W 1987 Value of growth hormone dynamics and somatomedin C (insulin-likegrowth factor I) levels in predictingthe long-term benefit aftertranssphenoidalsurgery for acromegaly. J Lab Clin Med 109:346 354

22. Biermasz NR, Smit JW, van Dulken H, Roelfsema F 2002 Postoperativepersistent thyrotrophin releasing hormone-induced growth hormone releasepredicts recurrence in patients with acromegaly. Clin Endocrinol (Oxf) 56:313319

23. Freda PU, Wardlaw SL, Post KD 1998 Long-term endocrinological follow-upevaluation in 115 patients who underwent transsphenoidal surgery for acro-megaly. J Neurosurg 89:353 358

24. Swearingen B, Barker 2nd FG, Katznelson L, Biller BM, Grinspoon S, Kli-banski A, Moayeri N, Black PM, Zervas NT 1998 Long-term mortality aftertranssphenoidal surgery and adjunctive therapy for acromegaly. J Clin En-docrinol Metab 83:3419 3426

25. Abosch A, Tyrrell JB, Lamborn KR, Hannegan LT, Applebury CB, Wilson

CB 1998 Transsphenoidal microsurgery for growth hormone-secreting pitu-itaryadenomas: initial outcomeand long-termresults.J ClinEndocrinol Metab83:34113418

26. Biermasz NR,van Dulken H, RoelfsemaF 2000 Ten-year follow-up results of transsphenoidal microsurgery in acromegaly. J Clin Endocrinol Metab 85:45964602

27. Kreutzer J, Vance ML, Lopes MB, Laws Jr ER 2001 Surgical management of GH-secreting pituitary adenomas: an outcome study using modern remissioncriteria. J Clin Endocrinol Metab 86:4072 4077

JCEM is published monthly by The Endocrine Society (http://www.endo-society .org), the foremost professional society serving theendocrine community.


Acromegalia 3

Documents

Transcript of Acromegalia 3