Acromegaly
Annamaria Colao,
Bartolomeo Merola,
Diego Ferone and
Gaetano Lombardi
Department of Molecular and Clinical Endocrinology and Oncology,
"Federico II" University, 80131 Naples, Italy
Address correspondence and requests for reprints to: Annamaria Colao, MD, PhD, Department of Molecular and Clinical Endocrinology and Oncology, "Federico II" University of Naples, via S. Pansini 5, 80131 Napoli, Italy.
 |
Introduction
|
---|
Acromegaly is a rare pituitary disorder
with an estimated incidence of three to four cases per million
population per year (1). Because it is a chronic and slowly developing
disease, clinically progressive disfigurements or disabilities go
unnoticed, and the diagnosis can be delayed. It is a severe systemic
disease, because the GH/insulin-like growth factor I (IGF-I) excess
causes impairment of cardiac and respiratory functions that contribute
to the increased mortality and morbidity (1).
This report refers to our clinical experience in the diagnosis
and management of acromegaly over a period of two decades, focusing
on specific clinical problems, such as cardiac morphology and function,
follow-up of pregnancy, and different approaches to therapy. During
these years we cared for 168 patients for a mean period of 8 yr (range
118 yr). Although approximately half of them were lost at follow-up
after 510 yr, and 11 died from different causes, the large number of
patients studied gave us the opportunity to draw a number of
preliminary conclusions.
 |
Cardiac features of acromegaly
|
---|
Because atherosclerosis, cardiovascular and cerebrovascular
diseases, and respiratory diseases double the death rate compared with
the healthy population, especially after the age of 45, additional
testing should be performed after the diagnosis of acromegaly is made.
In particular, we have developed a careful cardiological assessment
strategy, including standard and 24-h Holter electrocardiogram (ECG),
echocardiography, and equilibrium radionuclide scintigraphy. In
acromegaly, cardiac enlargement is a consistent finding and seems to be
disproportionate as compared with the increase in size of other
internal body organs (2). An increased frequency of cardiovascular
diseases, such as systemic hypertension, premature coronary disease,
arrhythmias, especially ventricular premature beats and
intraventricular conduction defects, and congestive heart failure have
been described (2). Most studies of acromegalic cardiopathy have
focused on structural and anatomical abnormalities of the heart (like
ventricular wall hypertrophy and/or ventricular dilatation).
Particularly, myocardial hypertrophy with interstitial fibrosis,
lymphomononuclear infiltration, and areas of monocyte necrosis
resembling myocarditis often resulting in increased left ventricular
mass (LVM) and concentric hypertrophy have been described (2). Similar
alterations have been observed for the right ventricle. Conversely,
little information is available on the diastolic function in
acromegaly, although it is known that diastolic abnormalities may
precede systolic dysfunction and may represent a distinct cause of
impaired cardiac function. Using Doppler echocardiography and
equilibrium radionuclide scintigraphy we have shown that abnormal
diastolic filling patterns of transmitral, transtricuspid, and superior
vena cava flowmetry are frequently present in acromegalic patients,
indicating an impaired relaxation associated with an increased left and
right ventricular mass (2). Moreover, cardiac performance was
significantly impaired in uncomplicated acromegaly as evidenced by the
significantly reduced ejection fraction during physical exercise as
compared with controls. This has been documented for both left (61
± 11% vs. 75 ± 8%, P < 0.001) and
right ventricle (45 ± 13 vs. 58 ± 11%,
P < 0.002). In 73% of our patients, the left
ventricular ejection fraction during exercise increased by less than
5% compared with basal values, thereby fulfilling the criteria for
impaired cardiac performance (2). These findings could indicate the
presence of a specific acromegalic cardiomyopathy, which seems to be
correlated with the duration of disease rather than with circulating GH
and/or IGF-I levels. Thus, acromegalics may be asymptomatic for many
years before showing clinical and/or echocardiographic features of
cardiac impairment. In the early stages of the disease, some patients
may even present with a hyperkinetic syndrome, characterized by
increased heart rate and cardiac output and decreased vascular
resistance. Frequently, left ventricular hypertrophy occurs first,
often leading to slow deterioration of diastolic function. Congestive
heart failure may develop when the disease is untreated or
unsuccessfully treated. Of 168 patients admitted to our Department for
acromegaly, only 3 developed heart failure as their initial symptom.
Two were elderly (65 and 67 yr old), whereas the third, a 37-yr-old
male, presented with an end-stage cardiac failure. Initial signs of
cardiac hypertrophy can also be recorded in young acromegalic patients
(<40 yr old) with disease duration shorter than 5 yr. With the aid of
monodimensional and pulsed Doppler echocardiography, we recently found
a significant increase of the LVM and the LVM indexed for body surface
area in 20 normotensive, untreated young acromegalics (Fig. 1
). In these patients both end-systolic
and end-diastolic dimensions, as well as the isovolumic relaxation
time, were significantly higher than controls (Fig. 1
) in the absence
of signs of cardiac impairment. In fact, the ejection fraction was
similar between patients and age- and sex-matched controls (67.6
± 2.5% vs. 66.6 ± 1.8%). These findings suggest an
early involvement of the cardiac muscle in young acromegalics as well.
The treatment with octreotide (OCT) lead to an improvement of cardiac
parameters. A significant decrease of LVM, interventricular septum
thickness, and right posterior wall thickness was found after 6 months
of OCT treatment (3), although hemodynamic parameters remained
unchanged after 12 months of treatment (4). Suppression of GH/IGF-I
levels for 24 months was not paralleled by a significant change in the
ejection fraction, either at rest or after exercise, in 11 acromegalics
treated with OCT (4). This indicates that a longer period of treatment
may be needed to normalize cardiac performance. It could be argued
that, during standard sc treatment with OCT, incomplete suppression of
the GH/IGF-I levels throughout the day might have contributed to the
failure to improve cardiac function. In agreement with this hypothesis
are our preliminary results in 3 patients treated for 18 months with
the long-lasting im formulation of OCT, at the dose of 20 mg/month
(Sandostatin LAR, Sandoz, Basel, Switzerland), showing a significant
decrease of the LVM and a significant increase of the ejection fraction
(in the 2 normotensive patients) after only 6 months of treatment.
 |
Follow-up of pregnancy in acromegalic women
|
---|
Pregnancy is a rather rare event in acromegalic women because
fertility is often reduced during the disease. Thus, only a few
pregnancies in acromegalic women have been reported in the literature.
We followed up on ten pregnancies in six patients with active
acromegaly (Table 1
). One pregnancy
spontaneously ended after 3 months, and the last one was ongoing at the
time of this report. In four women, acromegaly was first diagnosed
during pregnancy, and all patients delivered naturally healthy newborns
whose height and weight were on or over the 97th percentile. All
infants were breast-fed and in good health. Three women became pregnant
13 yr after surgery, and in two of them, pregnancy occurred when GH
and IGF-I levels were still slightly elevated. In two other cases
pregnancy occurred during chronic OCT treatment. OCT withdrawal was
recommended to these patients when pregnancy was confirmed. However,
one patient refused to discontinue OCT treatment (no.5, Table 1
)
because of persistent and analgesic-resistant headache; she continued
therapy at high doses (9001200 µg/day divided in multiple
administrations) during the whole period of pregnancy. In four women,
serum GH and plasma IGF-I levels were assessed every month, while
assessment of the GH secretory profile, with sampling every hour from
0800 to 1600 h, was carried out every 3 months. Circulating GH and
IGF-I levels were only suppressed in the patient who continued OCT
therapy. The patient who stopped OCT treatment gave birth to an
overweight girl (4.5 kg), while the patient who continued OCT treatment
had a girl whose weight was normal (3.2 kg). Against our
recommendations, the latter patient breast-fed her baby for 4 months
with no apparent problem. No growth of pituitary adenoma in the six
patients was shown by computed tomography (CT) or magnetic resonance
imaging (MRI) studies performed after delivery. Visual field studies
performed during pregnancy failed to reveal any defect. The two
patients with mildly elevated GH/IGF-I levels before pregnancy (nos. 3
and 6, Table 1
) reported an improvement of signs and symptoms during
pregnancy.
 |
Treatment of acromegaly
|
---|
The treatment of acromegaly is aimed at removing the source of GH
hypersecretion or at suppressing its activity. The effectiveness of
therapy is documented by the suppression of GH to levels below 2 µg/L
after glucose load and the normalization of IGF-I levels (5). Surgery
is considered the first choice of treatment, followed by radiotherapy
and/or medical therapy, on the basis of the presence and invasiveness
of tumor remnant (6). Among the 168 patients admitted to our Department
for acromegaly, 145 were treated by surgery. In 5 out of these 145, a
second operation was performed after 510 yr because of tumor
regrowth, as documented by CT and/or MRI. Because the cut-off levels of
serum GH to indicate cure has significantly changed during the years
from 10 to 2.5 µg/L, and because presently the normalization of
plasma IGF-I should also be considered, the number of patients cured by
surgery decreased progressively, with a total success rate of 40%.
Moreover, radiotherapy was used in 33 patients of our series, more
frequently in the past for the inadequacy of medical
treatment. Surgery together with radiotherapy caused the
cure of the disease in 19 out of 33 patients. It should be pointed out
that the majority of these 19 patients have demonstrated GH deficiency
at recent follow-ups. Taking into account the evidence that GH
deficiency in adult age is associated with metabolic and body
composition alterations and with impairment of cardiovascular function
(8), this finding may be clinically relevant. Table 2
summarizes the results of different
therapeutic approaches in our acromegalic patients during these last 20
years.
Medical treatment has greatly improved in the last decade with the
introduction in the clinical practice of somatostatin analogs, such as
octreotide (OCT) and lanreotide, and new dopamine agonists, such as
quinagolide (CV 205502) and cabergoline. Most studies indicate that
somatostatin analogs are more effective than dopamine agonists (1, 2, 7). We have recently shown that the treatment with quinagolide, at the
dose 0.30.6 mg/day for 6 months, normalized circulating GH and IGF-I
levels in only 43.8% of patients (9). The long-lasting im formulations
of bromocriptine and cabergoline were even less effective (9). OCT is
the most widely used drug, but because it must be administered at least
three times daily sc, the frequency of injections has been reported to
cause poor compliance and/or early side effects, sometimes leading to
therapy withdrawal (10). To test for tolerability and to predict
chronic responsiveness to treatment, the acute test with 50100 µg
OCT is commonly used (10). However, in our experience the acute OCT
administration did not prove to be a reliable pretreatment screening
test because of a sensitivity of 71% and a specificity of 55% (11).
The recent availability of lanreotide prompted us to test its
effectiveness in acromegalic patients and to compare it with the
effectiveness of OCT. In the first 15 patients treated with either one
of the drugs, both drugs strongly suppressed the GH/IGF-I
hypersecretion (Fig. 2
). However,
patients compliance was markedly better during lanreotide treatment
than during OCT treatment. In these patients, OCT caused a
significantly greater suppression of GH levels than lanreotide
(89.3 ± 3% vs. 77 ± 3.9%, P <
0.001, Fig. 2
). Three patients showing a poor response to OCT showed a
poor response to lanreotide treatment as well.

View larger version (21K):
[in this window]
[in a new window]
|
Figure 2. Serum GH (top) and IGF-I
(bottom) levels during octreotide and lanreotide
treatment in 15 patients (left) and corresponding
percent hormone suppression (right). All 15 patients
were treated first with OCT at a dose of 0.30.6 mg/day, in three
daily doses for 6 months and subsequently, after 715 day withdrawal,
with LAN, at the dose of 6090 mg/month for 6 months.
|
|
Although it is current opinion that pharmacotherapy should be
instituted after unsuccessful surgery (5, 6), OCT therapy before
surgery has been reported to improve the surgical outcome. In fact, the
treatment with OCT has been shown to improve glucose tolerance or
diabetes mellitus (5, 6, 9), cardiovascular parameters (2, 3, 4), and to
cause tumor shrinkage. Clearly, the achievement of improved metabolic
conditions is favorable for the anesthesiological procedure, while the
reduction of tumor mass can facilitate the neurosurgical excision.
Based on these observations, we administered OCT for 36 months before
surgery in a group of 22 acromegalics. OCT pretreatment caused a
significant decrease of serum GH in these 22 patients. GH and IGF-I
normalized in 12 OCT-pretreated patients (54.5%), between 1015 days
after surgery. GH and IGF-I normalized in only 11 of 37 patients who
did not receive OCT pretreatment (29.7%). The surgical outcome was
significantly improved in OCT-pretreated patients (P <
0.005,
2 test). In 3 out of 7 diabetic acromegalics
receiving OCT, glucose lowering drugs could be withdrawn because blood
glucose normalized on a low carbohydrate diet alone. In 2 diabetic
acromegalics, insulin could be replaced by oral glucose lowering drugs,
while in the remaining 2 patients the insulin dose could be reduced. In
OCT-treated patients the average blood glucose levels before surgery
were significantly lower than in untreated patients, and the levels
remained low at the first follow-up after surgery (Fig. 3
). Similarly, both circulating total
cholesterol and triglycerides levels were significantly higher in
untreated than in OCT-treated patients either before or after surgery
(Fig. 3
). In addition, both systolic and diastolic blood pressures
decreased in 5 patients, and ECG recording normalized in 7 of 11
OCT-treated patients. A significant tumor shrinkage was documented with
CT and/or MRI in 5 out of 22 OCT-treated patients and in none of the
untreated patients. Macroscopically, no difference was found between
untreated and OCT-pretreated adenomas. The tumor could be easily
removed in most cases (83.2% vs. 81.8%), and tumor
invasion of perisellar tissues was noticeable in 27% and 22.7% of
untreated vs. OCT-pretreated adenomas, respectively.
Pathology showed a significant increase of cellular atypia in
OCT-pretreated vs. untreated adenomas (31.6% vs.
19.2%, P < 0.05). Finally, from the analysis of the
results of this retrospective study it appeared that the time needed to
recover from surgery was significantly shorter in patients who were
pretreated with OCT than in untreated patients (5.6 ± 0.5
vs. 8.6 ± 0.7 days; P < 0.002). In
conclusion, a short-term treatment with OCT before surgery might
improve cardiac and metabolic conditions in acromegalics, reducing the
generic anesthesiological risk while also improving the clinical
recovery after surgery.
When OCT alone is ineffective in normalizing GH and/or IGF-I
concentrations, a combined treatment with dopamine agonists can be
considered, because the combined administration of these compounds has
been reported to be more effective in lowering GH levels compared with
either drug given alone. We have demonstrated that the combined
treatment with OCT plus quinagolide was effective in a few
therapy-resistant acromegalics (12). This effect is likely the result
of an increased bioavailability of dopamine agonists, as shown for
bromocriptine (12), caused by the combined administration with OCT.
 |
Conclusions
|
---|
Our Department serves an area of about 4 million people.
Therefore, the number of patients we have observed in the last 10 years
fits with the expected frequency of 34 cases/million population/year.
Of the 168 patients followed by our Department, 58 were cured by
surgery, 19 were cured after surgery and radiotherapy, and 91 had their
GH/IGF-I suppressed by medical treatment. Our studies on cardiac
function led us to conclude that the cardiovascular system can be
involved in the early stage of the disease. Successful treatment with
long-acting somatostatin analogs can prevent the occurrence of cardiac
disorders and can ameliorate cardiac function parameters as well. It is
noteworthy that pregnancy in acromegalic women does not seem to
significantly influence GH/IGF-I levels or tumor size. Apparently,
treatment had no adverse effect on the outcome of pregnancy, but
further studies are needed to firmly establish whether pregnant
acromegalic women need to be treated or not. Chronic administration of
somatostatin analogs represents the milestone of medical treatment of
acromegaly. We have also shown that a short-term course of OCT before
surgery can improve the surgical outcome. Clearly, the new long-acting
formulations represent a further advancement in GH suppression activity
as well as patients compliance. It is expected that in future years,
new more effective, and safer drugs will be available. By analogy with
the effectiveness of medical treatment of prolactinomas, it is hoped
that these drugs might become the treatment of choice for
acromegaly.
 |
Acknowledgments
|
---|
This paper is dedicated to the memory of Marco Minozzi, Chief of
the Endocrine Unit at the "Federico II" University of Naples, from
1973 to 1981.
We are greatly indebted to Paolo Marzullo for his skillful contribution
in taking care of our patients and to Sandro Loche for kindly revising
the manuscript.
Received July 31, 1996.
Revised May 29, 1997.
Accepted June 10, 1997.
 |
References
|
---|
-
Melmed S. 1990 Acromegaly. N Engl J
Med. 322:966977.[Medline]
-
Saccà L, Cittadini A, Fazio F. 1994 Growth
hormone and the heart. Endocr Rev. 15:555573.[Abstract]
-
Merola B, Cittadini A, Colao A, et al. 1993 Chronic treatment with the somatostatin analog octreotide improves
cardiac abnormalities in acromegaly. J Clin Endocrinol Metab. 77:790793.[Abstract]
-
Lombardi G, Colao A, Ferone D, et al. 1996 Cardiovascular aspects in acromegaly: effects of treatment. Metabolism.
45(Suppl.1):5760.
-
Melmed S, Dowling R, Frohman L, et al. 1994 Consensus statement: benefits versus risks of medical
therapy for acromegaly. Am J Med. 97:468473.[Medline]
-
Frohman LA. 1991 Therapeutic options in
acromegaly. J Clin Endocrinol Metab. 72:11751181.[Medline]
-
De Boer H, Block G-J, Van Der Veen EA. 1995 Clinical aspects of growth hormone deficiency in adults. Endocr Rev. 16:6386.[Medline]
-
Colao A, Ferone D, Marzullo P, et al. 1997 Effect
of different dopaminergic agents in the treatment of acromegaly. J
Clin Endocrinol Metab. 82:518523.[Abstract/Free Full Text]
-
Lamberts SWJ, van der Lely A-J, de Herder WW, Hofland
LJ. 1996 Octreotide. N Engl J Med. 334:246254.[Free Full Text]
-
Colao A, Ferone D, Lastoria S, et al. 1996 Prediction of efficacy of octreotide therapy in patients with
acromegaly. J Clin Endocrinol Metab. 81:13561362.
-
Lombardi G, Colao A, Ferone D, et al. 1995 CV
205502 treatment in therapy-resistant acromegalic patients. Eur J
Endocrinol. 132:559564.[Medline]
-
Fløgstad AK, Halse J, Grass P, et al. 1994 A
comparison of octreotide, bromocriptine, or a combination of both drugs
in acromegaly. J Clin Endocrinol Metab. 79:461465.[Abstract]