|
|
||||||||
CLINICAL STUDY |
Departments of Endocrinology and Metabolism and 1 Cardiology, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands
(Correspondence should be addressed to AM Pereira, Email: A.M.Pereira{at}lumc.nl)
| Abstract |
|---|
|
|
|---|
Patients and methods: We compared LV systolic and diastolic function in inactive patients with acromegaly (n = 22), who were divided into patients with long-term control by octreotide (n = 14) and patients with long-term cure by surgery/radiotherapy (n = 8). We also assessed these parameters in patients with active acromegaly (n = 17).
Results: In patients with active acromegaly, systolic function at rest was decreased by 18% (P < 0.01), LV mass index increased by 40% (P < 0.04) and isovolumetric relaxation time increased by 19% (P < 0.01), compared with patients with inactive acromegaly. These parameters were not different between well-controlled and cured patients. Using tissue Doppler imaging, the ratio between early and late diastolic velocity (E'/A' ratio) was decreased in active, compared with inactive acromegaly (0.75±0.07 versus 1.24±0.15; P < 0.01). This E'/A' ratio was considerably higher in cured, compared with octreotide-treated, patients (1.75±0.41 versus 1.05±0.1; P < 0.01).
Conclusion: Diastolic function is persistently and significantly more impaired in acromegalic patients with long-term control by octreotide than in surgically cured patients, which points to biological effects of subtle abnormalities in growth hormone secretion. Criteria for strict biochemical control of acromegaly should thus be reconsidered.
| Introduction |
|---|
|
|
|---|
Adequate treatment of GH excess can arrest, and even reverse, several of these cardiac changes. A total of 19 studies evaluated the effect of long-term suppression that is, 6 months or more of GH excess on cardiac function, summarized in Table 1
. From these studies it is evident that LV mass decreases in association with improved systolic and diastolic function in patients, in whom GH excess is well-controlled. Nonetheless, it is not entirely clear to which extent long-term successful biochemical control of GH excess can reverse cardiac function. For instance, although octreotide treatment improved LV ejection fraction (LVEF), measured after 1 year of treatment, LVEF did not normalize completely (8). Only one study compared cardiac function 5 years after normalization of GH/insulin-like growth factor I (IGF-I) excess and compared patients with controlled disease and patients with cured disease. There were no differences in cardiac function, assessed by radionuclide ventriculography, between patients who were well controlled by octreotide and those cured by surgery (13). However, all non-invasive techniques have major pitfalls in so far as they cannot measure LV pressures directly.
|
Therefore, we investigated whether there were differences in cardiac parameters between patients with long-term control of GH excess by treatment with octreotide and patients cured by surgery, using a group of patients with active acromegaly as a reference group. We used echocardiography including tissue Doppler imaging (TDI), which allows for a detailed and quantitative assessment of cardiac parameters including diastolic and systolic function (17).
| Materials and methods |
|---|
|
|
|---|
We studied 39 consecutive patients with acromegaly (19 men) referred from the outpatient clinic of the Leiden University Medical Center, Leiden, The Netherlands (Table 2
). The mean age of the patients was 56 years (range 2083 years). The diagnosis of acromegaly was based on the characteristic clinical features and confirmed by insufficient suppression of GH during a glucose tolerance test (GH nadir below 0.5 µg/l), and the presence of a pituitary adenoma on radiological imaging.
|
7.0 mM. None of the female patients were pregnant during the 9 months following echocardiography. The duration of disease was defined by the onset of clinical symptoms related to GH excess (carpal tunnel syndrome, sleep apnoea and arthralgias), and by careful comparison of old photographs. The end of disease duration was defined as the time of successful treatment (that is, cured or well-controlled disease). The duration of well-controlled disease or cure was defined as the time of successful medical treatment and/or transsphenoidal surgery with or without adjuvant radiotherapy, until the time of echocardiography. The local institutional ethics committee approved the study, and written informed consent was obtained from all subjects.
Echocardiography, data acquisition
Echocardiography was performed with the patients in the left lateral decubitus position using a commercially available system (Vingmed system FiVe/Vivid-7; General Electric Vingmed, Milwaukee, WI, USA). Standard parasternal (long and short axis) and apical views (two-, four- and five-chamber) were obtained. Standard continuous-wave and pulsed-wave Doppler examinations were performed. M-mode images were obtained from the parasternal long-axis views for quantitative assessment of LV dimensions, fractional shortening (FS) and LVEF (10, 18). LV mass was calculated by the cube formula, and using the correction formula proposed by Devereux et al. (19; see also 20):
![]() |
LVEDD, PWD and IVSD are defined in Table 3
(below). LV mass index (LVMI) was corrected for body surface area. LV hypertrophy was defined as a LVMI value above 135 g/m2 for men and 110 g/m2 for women. Systolic function was evaluated by measurements of FS and LVEF. The following parameters of diastolic function were measured: diastolic transmitral peak velocities (E and A wave), the E/A ratio, the isovolumetric relaxation time (IVRT) and the E-deceleration time. Quantitative diastolic data were derived from TDI analysis. For TDI analysis, the digital cineloops were analysed using commercial software (Echopac 6.1; General Electric Vingmed). The sample volume (4 mm3) was placed in the LV basal portion of the septum (using the four-chamber images). The following parameters (mean values calculated from three consecutive beats) were derived: early diastolic velocity (E'), late diastolic velocity (A') and the E'/A' ratio. All echocardiographic examinations and analyses were performed by a single observer, blinded for treatment modalities.
|
GH concentrations were quantitated in duplicate using a sensitive time-resolved immunofluorescence assay (Wallac, Turku, Finland), specific for 22 kDa GH protein, and calibrated against WHO IRP 80/505. The detection limit was 0.012 µg/l. Intra-assay coefficients of variation were 1.68.4% in the GH range 0.01218 µg/l. The total serum IGF-I concentration was determined by RIA after extraction and purification on ODS-silica columns (Incstar Corp., Stillwater, MN, USA). The intra- and inter-assay coefficients of variation were less than 11%. The detection limit was 1.5 nM. Age-related normal data were determined in the same laboratory. IGF-I was also expressed as a SDS from age-related normal levels.
Statistical analysis
Univariate analysis of variance was performed to compare groups, and the Bonferroni multiple comparison as a post hoc test. Linear-by-linear association was performed to investigate a trend for having untreated, uncontrolled, well-controlled or cured disease with an E'/A' ratio of < 1. Data are expressed as means±S.E.M. SPSS software version 11.0 (SPSS, Chicago, IL, USA) was used. Differences were considered statistically significant at the P < 0.05 level.
| Results |
|---|
|
|
|---|
The clinical characteristics are provided in Table 2
. GH and IGF-I concentrations were much higher in patients with active acromegaly, compared with the values obtained in patients with inactive acromegaly, reflecting the different inclusion criteria for the different groups. However, there were no differences in GH/IGF-I levels between untreated and uncontrolled patients, nor between well-controlled and cured patients. The duration of controlled GH/IGF-I levels was not different between well-controlled patients and cured patients (mean 5.8 years, range 114 years, versus mean 7.9 years, range 216 years, respectively; not significant).
LV dimensions and systolic function at rest
LV dimensions were not different between patients with active and inactive acromegaly (Table 3
). However, LVMI was above the normal range and 40% higher in patients with active acromegaly, compared with patients with inactive acromegaly (140±17.9 versus 99.8±8.8 g/m2, respectively; P < 0.04). LVMI was within the normal range and not different between well-controlled and cured patients.
Systolic function at rest, reflected by FS and LVEF, was decreased by 1819% in patients with active acromegaly, compared with patients with inactive acromegaly. FS was 30.3±1.8 versus 37.0±1.2%, respectively (P < 0.01) and LVEF was 58.8±2.3 versus 72.6±1.8%, respectively (P < 0.01). However, FS and LVEF were not different between well-controlled and cured patients.
Diastolic function
There were no statistically significant differences in diastolic transmural peak velocities (E and A waves) between patients with active and inactive acromegaly. The IVRT was increased by 19% in patients with active acromegaly compared with patients with inactive acromegaly (109.7±4.0 versus 88.7±2.5 ms; P < 0.01). However, there were no significant differences between untreated and uncontrolled patients, nor between well-controlled and cured patients.
Diastolic function, assessed by TDI (Fig. 1
), showed that the early diastolic velocity (E') was significantly higher in cured patients as compared with the other patient groups (Table 4
and Fig. 1
). A significant difference in E' was also noted between uncontrolled and well-controlled patients: E' was higher in well-controlled patients (P < 0.01), but was still significantly lower than in those who were cured of disease (P < 0.04). The E'/A' ratio was considerably decreased in patients with active acromegaly compared with patients with inactive acromegaly (0.75±0.07 versus 1.24±0.15; P < 0.01). In cured patients, the E'/A' ratio was significantly higher when compared with well-controlled patients (1.75±0.41 versus 1.05±0.1, respectively; P < 0.01). Remarkably, the E'/A' ratio was < 1 in all untreated patients and in 75% of uncontrolled patients. The E'/A' ratio was < 1 in 50% of the well-controlled patients compared with only 12% of the cured patients (P = 0.003).
|
|
| Discussion |
|---|
|
|
|---|
The question arises as to whether other factors may have affected our observations, other than these related to the disease activityof acromegaly. First, the duration of GH excess is a determinant of cardiac abnormalities (6, 8). The group characterized as having active disease had a longer duration of disease than the cured patients, which will have affected cardiac function. However, there were no significant differences in disease duration between well-controlled and cured patients. Second, the degree of GH excess is a determinant of cardiac abnormalities. However, there were no differences in GH levels obtained over several hours or IGF-I levels between well-controlled and cured patients. Third, there were no differences in duration of strict control between both groups. Fourth, there were no differences in body mass index or age, which may have affected our conclusions. Fifth, treated hypertension, which may have induced diastolic dysfunction, was present in 38% (3/8) of the cured patients but in none of the well-controlled group. Moreover, all patients with treated hypertension had a blood pressure of < 140/90 mmHg during the year prior to the study. Finally, the prevalence of diabetes mellitus and impaired glucose tolerance was not different between cured and well-controlled patients (25 versus 21%, respectively). Therefore, our observations are not affected by differences in blood pressure or carbohydrate metabolism. Based on these arguments, we feel that it is unlikely that our interpretation of the data is confounded by parameters other than those related to disease activity of acromegaly. However, we cannot exclude the possibility that the persistent cardiac impairment could be due to still-unknown factors other than GH hypersecretion, like asymptomatic ischaemia.
Treatment of GH excess favourably affects cardiac function and mass. To our knowledge, a total of 19 studies involving 312 acromegalic patients have been published, which have assessed the effect of treatment on cardiac function (Table 1
). Treatment of GH excess decreased LV mass and improved diastolic function invariably, whereas systolic function at rest remained unchanged in most of the studies. Our data are in accordance with these conclusions from the other studies. Of the 312 patients, 53% (166 patients) achieved treatment goals, defined by normalization of IGF-I and fasting GH levels below 2.5 µg/l or glucose-suppressed GH levels below 1 µg/l. The majority of these patients, 120/166 (72%), were well-controlled by somatostatin analogue treatment. In these 120 patients, LV mass normalized and cardiac function improved. This was reflected mainly by an increase in LVEF during exercise (2129), a feature that has already been observed within a few months of treatment with somatostatin analogues. Prolonged suppression of basal or glucose-suppressed GH levels to values below 2.5 or 1 µg/l, respectively, in combination with normalization of plasma IGF-I levels for at least 1 year, resulted in significant improvement, but not complete normalization, of LVEF either at rest or at peak exercise without significant changes in diastolic filling (24). These data suggest that prolonged suppression of circulating GH and IGF-I levels normalizes systolic cardiac performance. 46 of the 166 patients (27%) with inactive acromegaly were biochemically cured by surgery and/or radiotherapy and were not treated with somatostatin analogues. Hradec et al. (18) demonstrated a clear beneficial effect of long-term cure on LVMI, but diastolic function was not assessed in that particular study. This beneficial effect of cure of GH excess on LVMI was confirmed in other studies, with a concomitant improvement, but not normalization, in diastolic function (28, 29). The biochemical criteria used in the studies on cardiac function in acromegaly are based on other studies, which have shown that these criteria are associated with a reversal of the increased risk for malignancies and mortality, associated with GH excess (18, 30, 31). However, it is unknown whether biochemical control of GH/IGF-I excess, according to these criteria, is also sufficient to normalize other GH-related morbidity, like acromegalic cardiomyopathy.
The findings in the current study, in patients with long-term control of GH excess (median of 6 years), demonstrated that two independent parameters of diastolic function, the E'/A' ratio and the IVRT, improved significantly, indicating ameliorated relaxation and decreased stiffness of the heart muscle (see also 32). However, a significantly higher E'/A' ratio was found in patients cured by surgery when compared with those well-controlled with long-term octreotide. Furthermore, Sicolo et al. (33) showed that in the presence of diastolic impairment the incomplete recovery of an adequate preload can affect systolic parameters during physical effort. Since systolic function was only measured at rest, it is possible that systolic function could still be impaired on effort and hence there might be a difference between those cured and those in remission. Therefore, these data suggest that acromegalic patients, well controlled according to stringent criteria, still reveal biological effects of slight GH overproduction. In accordance, there are indications that treatment of active acromegaly with somatostatin analogues resulting in normal IGF-I and GH levels does not completely normalize GH secretion. Recently, we investigated 24-h GH profiles in uncontrolled and well-controlled acromegaly patients, treated with long-acting somatostatin analogues (34). We applied the same strict biochemical criteria for well-controlled disease (normal IGF-I levels and a GH profile during 24-h GH sampling of < 2.5 µg/l) in both groups. However, GH was actually sporadically below 0.5 µg/l, although chronic treatment with somatostatin analogues repressed amplitude-dependent measures of excessive GH secretion in acromegaly. Moreover, tumoral endocrine autonomy was inferred by continued elevations of event frequency, overall pattern disruption (irregularity), and nonsuppressible basal GH secretion. We postulate that these subtle abnormalities in GH secretion relate to the persistently impaired diastolic function despite clinically normal GH and IGF-I levels.
In conclusion, long-term control of GH/IGF-I excess is associated with normal LV mass and LV dimensions. Nonetheless, diastolic function is more impaired in well-controlled patients than in surgically cured patients, which proves that the current criteria for strict biochemical control of acromegaly may still be associated with subtle effects of excessive GH secretion. Although the clinical relevance of this observation remains to be determined, these patients might benefit from more agressive control of GH production than obtained by applying the current strict criteria of biochemical control of GH excess.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
O. M. Dekkers, N. R. Biermasz, A. M. Pereira, J. A. Romijn, and J. P. Vandenbroucke Mortality in Acromegaly: A Metaanalysis J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 61 - 67. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. L. Ronchi, V. Varca, P. Beck-Peccoz, E. Orsi, F. Donadio, A. Baccarelli, C. Giavoli, E. Ferrante, A. Lania, A. Spada, et al. Comparison between Six-Year Therapy with Long-Acting Somatostatin Analogs and Successful Surgery in Acromegaly: Effects on Cardiovascular Risk Factors J. Clin. Endocrinol. Metab., January 1, 2006; 91(1): 121 - 128. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |