Long-term Changes in Serum IGF-1 Levels After
Successful Surgical Treatment of Growth
Hormone–Secreting Pituitary Adenoma
BACKGROUND: Successful treatment of acromegaly is known to normalize serum
insulin-like growth factor 1 (IGF-1) levels within days after surgery. However, our clinical
observations indicate that many cases of acromegaly show delayed normalization of
serum IGF-1 levels after complete tumor resection.
OBJECTIVE: To study long-term changes of the serum IGF-1 levels in acromegalic
patients for whom surgical treatment was thought to be successful.
METHODS: A retrospective observational study was performed with 46 acromegalic
patients with no residual tumor on sellar magnetic resonance imaging, and a nadir
growth hormone of le ss than 0.4 mg/L on a postoperative ora l glucose tolerance test.
RESULTS: In all patients, serum IGF-1 levels returned to the normal reference values for
age and sex during the observational period (12-132 months). The mean duration from
the time of surgery until IGF-1 normalization was 10 months (range, 3 days-57 months).
Twenty-seven patients (59%) reached normal IGF-1 ranges within 3 months of surger y,
whereas 19 patients (41%) experienced delayed (.3 months) IGF-1 normalization.
Eleven patients (24%) recovered normal IGF-1 levels 12 to 57 months after surgery. The
possibility of delayed IGF-1 cure was increased 8.8-fold with an immediate postoperative
IGF-1 level increase of 100 mg/L.
CONCLUSION: Satisfactory remission of acromegaly by IGF-1 criteria was delayed in
a large proportion of acromegalic patients, especially those with high postoperative
IGF-1 levels. Hence, additional treatment can be delayed in clinically stable acromegalic
patients who show no evidence of residual tumors on postoperative magnetic reso-
nance imaging and a normal growth hormone suppressive response to a glucose load.
KEY WORDS: Acromegaly, Growth hormone, Insulin-like growth factor 1, Pituitary tumor, Surgery
Neurosurgery 73:473–479, 2013 DOI: 10.1227/01.neu.0000431480.87160.84 www.neurosurgery-online.com
A
cromegaly requires careful clinical atten-
tion, as it is associated with increased
morbidity and mortality resulting from
prolonged exposure to excessive growth hormone
(GH). As more than 95% of acromegalic patients
have a GH-secreting benign pituitary adenoma,
transsphenoidal adenectomy (TSA) remains first-
line treatment of acromegaly in many countries
despite the development of medical treatments
and radiation therapy. After TSA, a careful mon-
itoring of hormone state is needed to evaluate
a cure or recurrence of acromegaly.
The biochemical cure for acromegaly is gener-
ally defined by the normalization of age- and sex-
matched IGF-1 levels and the suppression of
serum GH levels to less than 0.4 mg/L in response
to a 75-g oral glucose load.
1
Given that the
elevation of serum IGF-1 levels is a sensitive and
specific indicator of persistent disease, the
normalization of IGF-1 is expected to occur
after a successful treatment. It has been reported
that serum IGF-1 levels return to a normal
reference value for age and sex within 3 to 4 days
after the complete removal of pituitary tumor
Mi-Seon Shin, MD*
Ji Hee Yu, MD*
Jong Han Choi, MD*
Chang Hee Jung, MD*
Jenie Yoonoo Hwang, MD*
Young Hyun Cho, MD‡
Chang Jin Kim, MD‡
Min-Seon Kim, MD, PhD*
*Division of Endocrinology and Metabo-
lism, Department of Internal Medicine;
‡Depar tment of Neurosurgery, Asan
Medical Center University of Ulsan Col-
lege of Medicine, Seoul, Korea
Correspondence:
Min-Seon Kim, MD, PhD,
Department of Internal Medicine,
Asan Medical Center,
University of Ulsan College of Medicine,
88, Olympic-ro 43-gil, Songpa-gu,
Seoul 138-736, Korea.
E-mail: mskim@amc.seoul.kr
Received, January 13, 2013.
Accepted, May 20, 2013.
Published Online, May 30, 2013.
Copyright ª 2013 by the
Congress of Neurological Surgeons
WHAT IS THIS BOX?
A QR Code is a matrix
barcode readable by QR
scanners,mobile phones
with cameras, and
smartphones. The QR
Code above links to
Supplemental Digital
Content from t his
article.
SANS LifeLong Learning and
NEUROSURGERY offer CME for subscribers
that complete questions about featured
articles. Questions are located on the SANS
website (http://sans.cns.o rg/). Please read
the featured article and then log into SANS
for this educational offering.
ABBREVIATIONS: GH, growth hormone; IGF-1,
insulin-like growth factor 1; OGTT, oral glucose
tolerance test;
SDS, SD score; TSA, transsphenoidal
adenectomy
Supplemental digital content is available for this article.
Direct URL citations appear in the printed text and are
provided in the HTML and PDF versions of this article on
the journal’s Web site (www.neurosurgery-online.com).
RESEARCH—HUMAN—CLINICAL STUDIES
TOPIC RESEARCH—HUMAN—CLINICAL STUDIES
NEUROSURGERY VOLUME 73 | NUMBER 3 | SEPTEMBER 2013 | 473
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
tissue.
2
However, previous studies have reported that IGF-1
levels may fluctuate during the immediate postoperative period
but then stabilize at around 3 months after surgery.
3,4
Several
studies have also suggested that on rare occasions, normalization
of IGF-1 levels may be delayed until 12 months or more after
surgery.
5-8
However, our clinical observations have indicated that serum
IGF-1 levels gradually decline and return to normal even longer
than 1 year after surgery in not an inconsiderable number of cases
of acromegaly. If this truly occurs in these patients, a diagnosis of
a cure of acromegaly based on the early postoperative IGF-1 levels
may be incorrect. Analysis of long-term changes in serum IGF-1
levels (.1 year after surgery) may provide useful information to
address this issue. Hence, our study was conducted to assess long-
term postoperative changes in the serum IGF -1 levels in
acromegalic patients with a normal GH suppression response
to a glucose load and no visible residual tumor.
SUBJECTS AND METHODS
Subjects
The medical records of 214 acromegalic patients who had been
diagnosed with GH-secreting pituitary adenoma and had undergone
TSA at the Asan Medical Center (Seoul, Korea) between January 2000
and January 2010 were reviewed retrospectively. The local ethics
committee approved this retrospective review protocol. Acromegaly was
diagnosed on the basis of biochemical findings (ie, a nadir GH level of $1
mg/L during a 75-g oral glucose tolerance test [OGTT], and elevated
serum IGF-1 levels relative to the normal ranges defined by a previous
Korean study) (see Table, Supplemental Digital Content 1, http://links.
lww.com/NEU/A556).
9
A diagnosis of acromegaly was supported by
clinical features such as acral enlargement, prognathism, hyperhidrosis,
headache, diabetes mellitus, hypertension, and the presence of pituitary
adenoma on magnetic resonance imaging (MRI). Finally, these diagnoses
were confirmed by histological examination of surgical specimens as GH-
producing pituitary adenoma. All patients underwent TSA performed by
the same experienced surgeon. The extent of resection was decided
intraoperatively, but tumor tissue was removed as much as possible.
Forty-six patients who received additional treatments such as long-acting
somatostatin analogs, long-acting dopamine agonists, conventional
radiotherapy, and gamma-knife radiosurgery, were excluded from the
study. Another 34 patients were excluded owing to their failure to attend
the clinic on a regular basis. Among the remaining 134 patients who
underwent TSA and received regular biochemical assessment for at least
1 year after surgery, 46 patients who might be considered to be surgically
cured were finally selected for the data analysis. In these cases, MRI failed
to identify visible residual tumors, and OGTT analysis identified a nadir
GH value of less than 0.4 mg/L (Figure 1).
Hormone Assays
Serum IGF-1 levels were measured before surgery (preoperative period)
and within 3 days after surgery (immediate postoperative period), and
at 3-month intervals for 1 year followed by once or twice yearly. Serum
IGF-1 levels were measured using a commercial sandwich immunoradio-
metric assay kit (Diagnostic Systems laboratories, Inc, Webster, Texas),
which is the same kit used in our reference data.
9
The sensitivity of this
test was 2 mg/L. The intra-assay and interassay coefficients of variation of
the test were 3.9% to 7% and 3.8% to 7.4%, respectively.
OGTT was performed preoperatively and within 3 days after surgery,
and then once yearly if the IGF-1 levels had not normalized. After
overnight fasting, 75 g of dissolved glucose was orally administered, and
blood was sampled every 30 minutes over 2 hours. Serum GH levels were
measured in duplicate using an immunoradiometric assay kit (TFB,
Tokyo, Japan). The sensitivity of this GH assay kit was 0.05 mg/L. The
intra-assay and interassay coefficients of variation of the test were 1.6%
to 2.4% and 1.0% to 2.8%, respectively.
Definition of Biochemical Cure
Following the most recent consensus statement,
1
a biochemical cure of
acromegaly is defined as age- and sex-matched normalization of serum
IGF-I levels and suppression of GH by an oral glucose load to less than
0.4 mg/L. Normal IGF-1 ranges for a typical Korean population were
derived from a previous report,
9
which indicated that these levels were
similar to those determined for other Asian
10
and white
11
populations.
Given that serum IGF-1 concentrations fluctuate and are influenced
by other factors,
12,13
an acromegaly cure based on IGF-1 criteria was
considered if the serum IGF-1 levels were within the normal range over
at least 2 consecutive measurements and persisted thereafter. Also, just a
1-time elevation of the IGF-1 level above normal was not considered a
recurrence or not cured. The subjects were categorized into 3 groups: early
cured (IGF-1 normalization within 3 months), delayed cured (IGF-1
normalization after 3 months), and not cured (failure to restore normal
IGF-1 levels during the follow-up period).
Radiological Evaluation
Sellar MRI was performed before surgery, at 3 to 6 months post-
surgery, and when relapse was suspected because of clinical symptoms or
biochemical study results. All participants were examined in the supine
position using a commercial 3-T MRI system (Achieva 3.0 T; Philips,
Best, the Netherlands) with a 3-mm thickness in routine images and
1.5-mm thickness in dynamic images. The MRI contrast was usually
pushed with a bolus, but was injected at 5 mL/s when a dynamic image was
required. The total volume of the contrast was 15 mL. On the basis of
preoperative MRI, an adenoma was classified as either micro- or
FIGURE 1. Flow chart of the current study population. GH, growth hormone,
MRI, magnetic resonance imaging.
SHIN ET AL
474 | VOLUME 73 | NUMBER 3 | SEPTEMBER 2013 www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
macroadenoma using a long axis cutoff value of 1 cm on coronal and
sagittal images. All MRI was reviewed by an experienced radiologist.
Statistical Analysis
Data were presented as the mean 6 SD for normally distributed
continuous variables, as medians and ranges for skewed-distributed
continuous variables, or as the number and proportion (%) for
categorical variables. The IGF-1 values was also presented as the SD
score (SDS), which enabled us to compare data with different normal
ranges. We calculated SDSs using the following formula:
SDS 5
ðX 2 mÞ
SD
;
X = measured IGF-1 value, m = the mean value of the reference
population, SD = the SD of the reference population. The IGF-1
reference values are shown in Supplemental Digital Content 1 (see
Table, http://links.lww.com/NEU/A556).
Demographic and biochemical characteristics between groups were
compared using a x
2
test for categorical variables and the Student t test or
Mann-Whitney U test for continuous variables. Logistic regression
analysis was used to identify clinical factors associated with delayed cure.
All statistical analyses were performed using the SPSS 17.0 software
(SPSS Inc, Chicago, Illinois). P , .05 was considered to denote
statistically significant differences.
RESULTS
Clinical and Biochemical Characteristics of the Subjects
Preoperative clinical and biochemical ch aracteristics of t he 46 acro-
megalic patients included in this study are presented in Table 1. The
mean age at the time of operation was 44 6 12 years. Both sexes
were almost equally represented. The median preoperative levels of
GH and IGF-1 were 20.2 mg/L (range, 2.4-830 mg/L) and 946.5
mg/L (range, 408-2270 mg/L) (SDS of 19.1, range, 13.3 to 124),
respectively. Twelve patients (26%) had microadenoma and 34
patients (74%) had macroadenoma. None of the patients had
systemic diseases such as uncontrolled diabetes, thyroid disease,
catabolic states, severe hepatic or renal disease, or malnutrition,
which might have affected the serum GH and IGF-1 levels.
Classification of Patients According to
IGF-1 Normalization
The mean follow-up duration after TSA was 61 months (range,
12-132 months). During this period, all patients achieved normal
IGF-1 levels. The patients who met the IGF-1 cure criteria were
again divided into the early-cured and the delayed-cured groups on
the basis of the time required for IGF-1 normalization, using 3
months as the cutoff. Twenty-seven patien ts (59%) were assigned
to the early-cured group and 19 patients (41%) to the delayed-
cured group (Figure 2A).
Time Taken to IGF-1 Normalization
The overall mean duration taken to IGF-1 normalization was
10 months (range, 3 days-57 months). When we analyzed the time
taken to IGF-1 normalization in 19 patients who belonged to the
delayed-cured group, the mean duration from surgery to IGF-1
normalization was 21 months (range, 4-57 months). Only 8
patients (17%) displayed normal IGF-1 levels within 12 months of
surgery, whereas 11 subjects (24%) showed normal IGF-1 levels
after 12 months (Figure 2B). Notably, 5 subjects (11%) recovered
normal IGF-1 levels after 2 years.
Comparisons of the Clinical and Biochemical
Characteristics Between Early- and Delayed-Cured and
Not-Cured Patients
The clinical and biochemical characteristics of the early- and
delayed-cured groups were compared (Table 2). The age and sex
were not different between the groups as were the preoperative
and immediate postoperative GH levels and preoperative IGF-1
levels. However, the immediate postoperative IGF-1 levels were
significantly higher in the delayed-cured group than in the early-
cured group. Moreover, the delayed-cured group had higher
frequency of macroadenoma (Table 2). However, changes in
IGF-1 levels before and after surgery and the longest tumor
diameter before surgery did not differ between the 2 groups.
In the early-cured group, 4 patients had type 2 diabetes mellitus
before surgery. One achieved a remission of diabetes and the other 3
reduced their antidiabetic medication (metformin plus sulfonylurea
or insulin to metformin alone) during follow-up. In the delayed-
cured group, 2 subjects had type 2 diabetes, which was well controlled
with metformin alone (glycated hemoglobin levels of ,7%) during
the observation period. Five in the early-cured group and 3 in the
delayed-cured subjects had hypertension and took antihypertensive
drugs. In addition, no fracture occurred in any patients. Therefore,
clinical complications of acromegaly in the delayed-cur ed group
were rather mild even if they had elevated IGF-1 levels.
Biochemical characteristics of delayed-cured subjects were also
compared with those who received additional treatments (n = 46)
and thus thought to be definitely not cured. Subjects in the
TABLE 1. Preoperative Clinical and Biochemical Characteristics of
the Patients
a
No. 46
Age, y 44 6 12
Sex, male:female 20:26
Random GH, mg/L 20.2 (2.4-830)
Random IGF-1, mg/L 946.5 (408-2270)
Random IGF-1, SDS 19.1 (13.3 to 124)
Tumor size
,1, no. (%) 12 (26)
$1, no. (%) 34 (74)
Long diameter, cm 1.3 (0.5-3.5)
a
GH, growth hormone; IGF-1, insulin-like growth factor; SDS, SD score. Data are
presented as the mean 6 SD, median and range, or number and proportion (%).
DELAYED IGF-1 CURE OF ACROMEGALY
NEURO
SURGERY VOLUME 73 | NUMBER 3 | SEPTEMBER 2013 | 475
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
delayed-cured group had lower preoperative and immediate
postoperative GH levels compared with those in the not-cured
group. The immediate postopera tive IGF-1 levels were also
significantly lower in the delayed-cured group than in the not-
cured group (Table 3). Moreover, the former group had a smaller
size of pituitary tumor than the latter group.
Facto rs Predicting a Delayed IGF-1 Cure
To identify factors that may help to predict a delayed IGF-1
cure, univariate and multivariate logistic regress ion analyses of
clinical, biochemical, and radiological parameters were performed.
The possibility of a delayed cure was increased by 8.8-fold (95%
confidence interval: 1.25-62.11, P = .029) compared with the
early-cured group, as the immediate-postoperative IGF-1 levels
increased by 100 mg/L. On the other hand, the probability of
a not cure over a delayed cure was increased by 2.2-fold (95%
confidence interval: 1.04-5.09, P = .041) as tumor size increased
by 1 cm. None of the other factors were helpful in predicting the
likelihood of either a delayed cure or a not cure.
DISCUSSION
Our current study presents 3 important clinical observations in
acromegalic patients who had undergone surgical treatment. First,
patients with only elevated IGF-1 levels but no evidence of residual
tumors on sellar MRI and normal GH suppression on OGTT
analysis could eventually recover normal IGF-1 levels. Second,
many of the cured acromegalic patients displayed elevated IGF-1
levels for a long period, which norm alized more than 1 year after
surgery without further treatment. Third, patien ts with high
immediate postoperative IGF-1 levels and a low rate of decrease of
IGF-1 after surgery had the higher probability of delayed cure.
Although the attainment of both normalized IGF-1 and nadir
GH levels suggests an absolute cure in patients with acromegaly,
discordant results between the 2 parameters have been reported by
many studies, with discrepancies noted in approximately 35% of
patients.
3,14-17
Consistent with these earlier reports, 46 of 134
cases in our study cohort (34%) showed elevat ed IGF-1 levels
during the early postoperative period despite normal GH
suppression to glucose loading. Moreover, 68 of 88 patients
(77%) who had a nadir GH of 0.4 mg/L or more on immediate
postoperative OGTT and/or residual tumor on sellar MRI finally
achieved a cure by IGF-1 criteria during the observation period.
Regarding the time taken for IGF-1 normalization after success-
ful surgery, it has been generally believed that IGF-1 levels may fall
into normal ranges within a few days. However, previous studies
suggested that serum IGF-1 levels slowly declined during the course
of the first 3 postoperative months.
3,4
Furthermore, Espinosa-
de-Ios-Monteros et al
3,8
showed that a significant number of
acromegalic patients may undergo biochemical changes on long-
term follow-up after surgery. In 8 of the 19 patients in their cohort
in the modified biochemical category, high IGF-1 levels became
FIGURE 2. A, classification of the acromegalic patients according to the time taken for insulin-like growth factor 1 (IGF-1)
normalization. Numbers and percentages of the patients who belong to the early, delayed, and non-cured groups.
B, numbers and
percentages of the patients who achieved IGF-1 normalization during the indicated period after surgery.
SHIN ET AL
476 | VOLUME 73 | NUMBER 3 | SEPTEMBER 2013 www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
normalized within 11.6 6 5.7 months post-surgery (range, 5-21
months).
3
Our current observation that the mean duration from
surgery to IGF-1 normalization was 10 months (range, 3 days-57
months) is in line with the previous report.
Freda
13
suggested that adolescence, pregnancy, or hyperthy-
roidism may falsely elevate the IGF-1 level, leading to discrepant
GH and IGF-1 values in patients with cured acromegaly.
However, none of our patients had these clinical situations.
TABLE 2. Comparisons of Clinical and Biochemical Characteristics of Early-Cured and Delayed-Cured Groups by Insulin-like Growth Factor 1
Criteria
a
Early-Cured (n = 27) Delayed-Cured (n = 19) P Value
Age, y
47 6 12 40 6 12 .082
Sex, male:female 14:13 6:13 .177
Random GH, mg/L
Preoperatively 15.6 (2.8-206) 23.2 (2.4-830) .647
Immediate postoperatively 1.2 (0.1-11.2) 1.6 (0.1-5.2) .964
Nadir GH, postoperatively, mg/L 0.2 (0.1-0.39) 0.3 (0.1-0.39) .103
Random IGF-1, mg/L
Preoperatively 910 (524-1728) 978 (408-2270) .157
Immediate postoperatively 334 (102-1274) 499 (223-1614) .024
Random IGF-1 (SDS)
Preoperatively 19.1 (15.4 to 117.9) 19.1 (13.3 to 124.0) .366
Immediate postoperatively 12.3 (21.2 to 112.4) 13.6 (10.9 to 116.3) .051
DIGF-1, mg/L
b
532 (305-723) 483 (185-1085) .384
DIGF-1, SDS 5.6 (2.0-18.2) 6.1 (0.7-12.6) .797
Preoperative tumor size
$1, no. (%) 17 (63) 17 (89) .046
Long diameter, cm 1 (0.7-2.5) 1.3 (0.5-3.5) .179
a
GH, growth hormone; IGF-1, insulin-like growth factor; SDS, SD score. Data are presented as the mean 6 SD, median and range, or number and proportion (%). Immediate
postoperatively is within 3 days after surgery.
b
DIGF-1 is the difference in preoperative IGF-1 levels and immediate postoperative IGF-1 levels.
TABLE 3. Comparisons of Clinical and Biochemical Characteristics of the Delayed-Cured Group by Insulin-like Growth Factor 1 Criteria and the
Not-Cured Group Who Had Additional Treatment
a
Delayed-Cured (n = 19) Not-Cured (n = 46) P Value
Age, y
40 6 12 41 6 11 .654
Sex, male:female 6:13 18:28 .569
Random GH, mg/L
Preoperatively
23.2 (2.4-830) 40.1 (4.8-305) .007
Immediate postoperatively 1.6 (0.1-5.2) 7.1 (0.1-174) .000
Nadir GH, postoperatively, mg/L 0.3 (0.1-0.39) 4.7 (0.68-48.1) .000
Random IGF-1, mg/L
Preoperatively 978 (408-2270) 1043 (406-2512) .299
Immediate postoperatively 499 (223-1614) 780 (30.1-1236) .031
Random IGF-1, SDS
Preoperatively 19.1 (13.3 to 124.0) 111.3 (13.4 to 127.8) .192
Immediate postoperatively 13.6 (10.9 to 116.3) 16.5 (22.5 to 111.9) .015
DIGF-1, mg/L
b
483 (185-1085) 334 (10-1953.9) .075
DIGF-1, SDS 6.1 (0.7 to 12.6) 4.2 (0.1 to 23.4) .182
Preoperative tumor size
$1, no. (%) 17 (89) 45 (98) .147
Long diameter, cm 1.3 (0.5-3.5) 2 (0.9-4.4) .021
a
GH, growth hormone; IGF-1, insulin-like growth factor; SDS, SD score. Data are presented as the mean 6 SD, median and range, or number and proportion (%). Immediate
postoperatively, within 3 days after surgery.
b
DIGF-1 is the difference in preoperative IGF-1 levels and immediate postoperative IGF-1 levels.
DELAYED IGF-1 CURE OF ACROMEGALY
NEURO
SURGERY VOLUME 73 | NUMBER 3 | SEPTEMBER 2013 | 477
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.