| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Department of Medicine (T.A.E., T.A.A., R.N.C.), School of Postgraduate Medicine, Keele University, Staffordshire, United Kingdom; Departments of Nuclear Medicine (J.O.) and Clinical Biochemistry (R.N.), North Staffordshire Hospital, Stoke-on-Trent, United Kingdom; and Section of Vascular Biology and Medicine (G.K., M.M., J.J.F.B.), University Department of Medicine, Ninewells Hospital and Medical School, Dundee, United Kingdom
Address all correspondence and requests for reprints to: Prof. R. N. Clayton, Department of Medicine, School of Postgraduate Medicine, University of Keele, Thornborow Drive, Stoke-on-Trent ST4 6QJ, United Kingdom. E-mail: r.n.clayton{at}keele.ac.uk
Abstract
Adult hypopituitarism is known to be associated with reduced life expectancy related to excess vascular events, and endothelial dysfunction is present in patients with this condition. We studied the relationship between biophysical and biochemical markers of endothelial dysfunction, including E-selectin, intercellular cell adhesion molecule-1, von Willebrand factor, and thrombomodulin in 52 adult patients with hypopituitarism and severe GH deficiency (<2 ng/ml on provocative testing) compared with 54 age-, sex-, and smoking-matched normal controls. We also examined endothelium-dependent dilatation of the brachial artery to postischemic occlusion and carotid artery morphology (intima-media thickness) by high-resolution ultrasonography. The patients were stable on conventional hormone replacement therapy but not on GH therapy, and none of the subjects had a known risk factor for vascular disease. Levels of E-selectin [57 ± 3 vs. 49 ± 2 ng/ml (mean ± SEM)] (P < 0.043), intercellular cell adhesion molecule-1 (308 ± 11 vs. 266 ± 10 ng/ml) (P < 0.001), thrombomodulin (49 ± 3 vs. 35 ± 2 ng/ml) (P < 0.001), and von Willebrand factor (132 ± 7% vs. 105 ± 5%) (P < 0.004) were significantly higher in patients than in controls. Brachial artery endothelium-dependent dilatation was significantly lower in patients than in controls [4.7% (0.00–9.77) vs. 10.5% (6.4–16.2) (median, interquartile range)] (P < 0.001). This difference in endothelium-dependent dilatation was more marked in female patients than in controls (P < 0.003), although it disappeared when estrogen-sufficient female patients were compared with controls (P = 0.31). However, the female patients who were not replaced with estrogen continued to show a striking difference compared with estrogen-deficient control females (P < 0.004). There was no difference in carotid intima-media thickness between patients of either sex and controls. On univariate analysis, brachial artery endothelium-dependent dilatation correlated inversely with intercellular cell adhesion molecule-1 (r = -0.225, P < 0.033). Intercellular cell adhesion molecule-1 correlated positively with E-selectin (r = 0.466, P < 0.0001) and negatively with IGF-I (r = -0.238, P < 0.016). E-selectin correlated with thrombomodulin (r = 0.215, P < 0.034) and von Willebrand factor (r = 0.218, P < 0.03) and negatively with IGF-I (r = -0.255, P < 009). Thrombomodulin correlated positively with von Willebrand factor (r = 0.422, P < 0.0001) and inversely with IGF-I (r = -0.266, P < 0.008). These correlations persisted after correction for age, sex, body mass index, and waist to hip ratio, with the exception of IGF-I, which now correlated with thrombomodulin only. These results confirm significant endothelial dysfunction in hypopituitarism and provide insight into the relationship of biochemical and biophysical markers of early atherosclerosis in hypopituitary GH-deficient adults. The negative correlation of IGF-I with some biochemical markers of endothelial dysfunction and the predictive nature of GH deficiency in stepwise regression analysis in this study supports the hypothesis that GH deficiency may play a role in these abnormalities. Future studies will determine whether GH treatment can reverse these abnormalities. Furthermore, the more significant endothelium-dependent dilatation abnormality in the female estrogen-deficient subjects compared with those who were estrogen replete suggests that estrogen replacement in these patients is a crucial element in protecting against vascular disease.
ADULT PATIENTS WITH hypopituitarism are known to have reduced life expectancy as a result of excess vascular events despite adequate conventional hormone replacement therapy in some (1, 2, 3) but not all epidemiological studies (4, 5). This reduced life expectancy is related to increased cardiovascular events (1, 2) as well as cerebrovascular accidents (2, 3). Premature atherosclerosis has been described in patients with hypopituitarism, as shown by increased intima-media thickness of carotid arteries (6, 7, 8). There is well-established evidence for clustering of various risk factors for vascular disease in these patients, including lipid abnormalities, visceral adiposity, glucose intolerance, insulin resistance, and hypertension (9). An early report has also suggested abnormalities of coagulation and fibrinolysis with increasing tendency to thrombosis (10).
Endothelial dysfunction is also thought to be an early event in the development of the atherosclerotic process (11). It is a known feature of diseases associated with increased vascular events, such as diabetes mellitus (12). Moreover, patients with asymptomatic atherosclerotic disease may have evidence of endothelial dysfunction (13), like subjects with known risk factors for vascular disease, such as cigarette smoking (14) and dyslipidemia (15). The vascular endothelium plays a vital and complex role in regulating both hemostasis and vascular tone. When activated, the vascular endothelium changes the balance between mechanisms that control thrombosis and vasoconstriction and those favoring vasodilation and fibrinolysis. Flow-mediated dilatation of the brachial artery is a surrogate for the endothelium-mediated responses generated by the release of nitric oxide from endothelial cells as a result of increased blood flow, causing shear stress at the endothelial cell surface (16). Thus, flow-mediated brachial artery dilatation is a well-documented measure of the functional state of the endothelium (17).
Various injurious stimuli mediate the generation of cytokines such as
TNF-
and IL-1 (18), and in response to these cytokines,
vascular endothelial cells express several glycoproteins, including the
adhesion molecules E-selectin and intercellular adhesion molecule-1
(ICAM-1) (19). These molecules facilitate the attachment
of macrophages and leukocytes and their migration to the subendothelial
space, where their products facilitate the atherogenic process
(20). This may show itself as increased intima-media
thickness (IMT), a well-defined early marker of premature
atherosclerosis (21). Increased IMT is directly related to
mortality from coronary artery disease (22). The activated
endothelium is also known to express other markers such as von
Willebrand factor (vWF) and thrombomodulin (TM).
To date, few studies have examined endothelial dysfunction and premature atherosclerosis in acquired hypopituitarism (8, 23, 24, 25), and none has examined the relationship between the biochemical and biophysical markers of endothelial dysfunction in patients with this condition. Accordingly, we studied markers of endothelial cell activation, including the soluble adhesion molecules E-selectin and ICAM-1, TM, and vWF, as well as functional abnormalities of the endothelium denoted by brachial artery dilatory response to postischemic stimulus and carotid artery morphology in patients with adult-onset hypopituitarism.
Subjects and Methods
Fifty-two patients with partial or complete hypopituitarism were recruited from the Endocrine Unit at North Staffordshire Hospitals NHS Trust between January 1998 and May 1999. All patients had severe GH deficiency, defined as maximum GH less than 5 mU/liter (<2 ng/ml) on an insulin stress test after adequate hypoglycemia (plasma glucose <2.2 mmol) or after a glucagon test if the insulin stress test was contraindicated, plus at least one other axis deficiency (except one patient). The causes of hypopituitarism were as follows: 26 patients had nonfunctioning pituitary adenoma, 8 had craniopharyngioma, 2 had parasellar meningioma, 2 had prolactinoma, 2 had Rathke’s cleft cyst, 2 had astrocytoma, and 1 patient each had optic nerve glioma, optic nerve germinoma, pituitary apoplexy, idiopathic hypopituitarism, suprasellar ectopic pinealoma, lymphocytic hypophysitis, sarcoidosis, suprasellar colloid cyst, olfactory nerve glioma, and hypothalamic hypopituitarism. Sixteen patients had panhypopituitarism with all anterior pituitary hormones and antidiuretic hormone deficiency (five-axis deficiency), 26 patients had a combination of GH, gonadal, adrenal, and thyroid axes deficiency (four-axis deficiency), 6 patients had intact pituitary-thyroid axis, 3 patients were not sex steroid deficient, and only 1 patient had isolated GH deficiency. Twenty-five patients were treated with surgery and radiotherapy, 21 patients were treated with surgery alone, 1 patient was treated with radiotherapy alone, and 1 patient was treated with surgery followed by radiotherapy and medical treatment (a dopamine agonist). Eight patients did not receive any treatment.
All patients were fully stabilized on conventional hormone replacement therapy with T4, sex steroids, and glucocorticoids where appropriate for at least 1 yr, but none was on GH treatment. None of the patients or controls was known to have a history of chronic inflammatory disorder or an acute inflammatory episode at the time of the study. Patients with acromegaly or Cushing’s disease were excluded.
There were 22 female and 30 male patients, age (mean ± SEM) was 47 ± 2 yr, body mass index (BMI) was 29 ± 1.0 kg/m2, waist to hip ratio (WHR) was 0.91 ± 0.02, and estimated duration of hypopituitarism and GH deficiency was 9 ± 1 yr. A cohort of 54 healthy normal volunteers recruited from hospital and academic staff as well as patient spouses and friends were studied as controls, of whom there were 24 females and 30 males, with age (mean ± SEM) of 46 ± 2 yr, BMI of 26 ± 1.0 kg/m2, and WHR of 0.85 ± 0.02.
There were 15 female controls and 13 female patients not on sex hormone replacement therapy, and 8 premenopausal women in the patient group and 5 in the control group were taking oral contraceptives. All male patients (except one) with pituitary-gonadal axis deficiency were on androgen replacement therapy. The North Staffordshire Medical Ethics Committee approved the study, and all participants gave informed consent to participate in the study.
The two groups were well matched for age, sex, and smoking status
(n = 9 in each group), and none of the patients or controls had
diabetes mellitus or impaired glucose tolerance (fasting glucose <6.1
mmol/liter), hypertension (blood pressure <150/90 mm Hg), or renal
disease (serum creatinine <120 µmol/liter). All patients and
controls were studied between 0830 and 0930 h in a fasting state,
and all underwent a full clinical examination including heart rate and
blood pressure measurement after a 10-min rest period. Baseline resting
electrocardiography and assessment of ankle brachial pressure index by
Doppler pressure ultrasonography to exclude occult peripheral arterial
occlusive disease (26) were conducted in all subjects
(Table 1
).
|
Cell adhesion molecules. Blood was centrifuged for 15 min at 4 C at 3500 rpm, and serum was removed and stored at -70 C. Levels of soluble serum E-selectin and ICAM-1 were measured with commercially available ELISA kits using the R & D system (Research and Development European Products Ltd., Abingdon, United Kingdom). The intraassay coefficients of variation for E-selectin and ICAM-1 assays were 1.0 and 2.7%, respectively, and the interassay coefficients of variation were 5.7% for E-selectin and 9.3% for ICAM-1.
vWF. vWF was measured in plasma by ELISA using DAKO Corp. (Glostrup, Denmark) reagent. The interassay coefficient of variation was 2.8%.
TM. TM was measured in plasma samples anticoagulated with 3.2% trisodium citrate 1:9. Blood samples were placed immediately in ice and centrifuged within 30 min at 4 C at 3500 rpm for 15 min. Plasma was separated and stored in aliquots at -70 C until assay. Plasma levels of TM were measured using an ELISA (Diagnostico Stago, Asnieres-Sur-Seine, France). The interassay coefficient of variation was 8.1%.
GH and IGF-I. GH concentration was determined using an automated two-site monoclonal/polyclonal chemiluminescent immunoassay on an Immulite analyzer (Diagnostic Products, Los Angeles, CA). IGF-I was measured with an immunoradiometric assay on acidified serum without extraction. The intraassay and interassay coefficients of variation were less than 5 and 4.5%, respectively.
Other biochemical tests. Serum cholesterol, triglycerides, and glucose were measured using commercial enzymatic methods (Randox Laboratories, Belfast, Northern Ireland) on an automated analyzer (DAX, Bayer Corp., Basingstoke, UK). The cholesterol content of high-density lipoproteins was determined using a direct nonprecipitation method (Wako Corp., Tokyo, Japan), and low-density lipoproteins were calculated using the Friedewald formula (27).
Biophysical measurements
Flow-mediated endothelium-dependent vasodilation. Endothelium-dependent vasodilation (EDD) was measured in the right brachial artery in each subject by the same observer. After 10 min of rest, brachial artery blood flow and the diameter of the brachial artery 2 cm above the antecubital fossa were determined using Doppler ultrasonography (Advanced Technology Labs, Bothell, WA) HDI 3000 digital ultrasound machine using a variable 5- to 10-megahertz signal transducer). To induce hyperemia, a standard sphygmomanometer cuff was applied to the forearm just proximal to the wrist joint and inflated to 300 mm Hg for 5 min. Blood flow and brachial artery diameter measurements were repeated 45–60 sec after the release of sphygmomanometer pressure. EDD and flow were calculated as the percentage change in brachial artery diameter and percentage change in flow, respectively (17).
Carotid IMT. Carotid IMT was determined by high-resolution ultrasonography 1 cm proximal to the right carotid bifurcation as the mean of three consecutive measurements of distal wall IMT, as described previously (8).
Ankle brachial pressure index. Ankle brachial pressure index at rest was estimated using an ultrasound Doppler flow detector (model 811-B, Parks Medical Electronics, Inc., Portland, OR). The measurements were taken according to the standard procedure reported previously (26).
Statistical analysis
Statistical analyses were performed using SPSS, Inc. (Chicago, IL). Variables were tested for normal distribution, and where confirmed, results are expressed as mean ± SEM for the biochemical data; otherwise, they are given as median and interquartile range for the biophysical data. Student’s t test was used to examine the difference between the normally distributed variables, and the Mann-Whitney U test was used for the other variables or where the number of subjects in groups was less than 15. Pearson’s correlation coefficient for biochemical data, Spearman’s correlation coefficient for biophysical data, and stepwise linear regression analysis were used to study the relationship between variables. A two-tailed P value less than 0.05 was taken as showing statistical significance.
Results
There was no difference in mean age between patients and controls;
however, the patient group had higher BMI and WHR compared with the
control group (29 ± 0.7 vs. 26 ± 0.5
kg/m2 and 0.91 ± 0.02 vs.
0.85 ± 0, respectively) (P < 0.001). When
analyzed by gender, BMI and WHR were significantly greater in patients
vs. controls (see Table 3). The ankle-brachial systolic
blood pressure index was similar in patients and controls. The levels
of various lipid parameters were different between patients and
controls, with total cholesterol greater in patients than in controls
(the difference largely related to increased low-density lipoprotein
subfraction). Likewise, triglyceride levels were also higher in
patients vs. controls. The difference in lipid parameters
was more marked in the female patient subgroup, and the details of
these are reported elsewhere (28). All patients and
controls had normal resting electrocardiograms. The demographic details
of the study groups are shown in Table 1
.
|
Levels of E-selectin, ICAM-1, vWF, and TM were significantly
higher in the patient group compared with the control group
(P = 0.043, 0.001, 0.004, and 0.001, respectively, by
Student’s t test) (Table 2).
|
) (r = -0.238, P
= 0.016) and positively with E-selectin (Fig. 1b) (r = 0.466,
P < 0.0001). E-selectin correlated positively with TM
(r = 0.215, P = 0.034) and vWF (r = 0.218,
P = 0.03) (Fig. 2) and
negatively with IGF-I (r = -0.255, P = 009) (Fig. 3). TM correlated positively with vWF
(r = 0.422, P = 0.001) and negatively with IGF-I
(r = -0.266, P = 0.008). The correlation of
ICAM-1 with E-selectin and that of TM with vWF persisted after
correction for the confounding effects of age, smoking, BMI, WHR, and
systolic blood pressure. After correction for the aforementioned
confounders, IGF-I still correlated negatively with TM but no longer
with E-selectin, ICAM-1, or vWF.
|
|
|
EDD was lower in GH-deficient patients compared with controls
(P = 0.001) (Table 2). Hyperemia-induced blood flow
increase was greater in patients than in controls, but this difference
was not statistically significant because of the wide variability
[94.2% (56.8–170.7) vs. 57.4% (39.2–130.5)]. There was
no difference in carotid IMT between patients and controls (Table 2).
On univariate analysis, brachial artery EDD correlated inversely with
ICAM-1 (r = -0.225, P < 0.033) (Fig. 4).
|
). ICAM-1 levels were
significantly higher in the female patients (estrogen sufficient and
deficient together) than controls, but not in the male patients
vs. controls. Although E-selectin was higher in the female
patients than controls, this difference was not significant. There were
similar findings for vWF and TM: both female and male patients had
higher values compared with female and male controls (Table 3).
Separating the female groups by estrogen status [estrogen sufficient
(intact or hormone replaced) or estrogen deficient], EDD was
significantly lower in the estrogen-deficient female patients (n =
11) compared with controls (n = 11). On the contrary, there was no
significant difference in EDD between estrogen-sufficient female
patients (n = 8) and their controls (n = 11) (Table 4). This difference persisted in
estrogen-sufficient GH-deficient females for ICAM-1, E-selectin, and TM
but not for vWF; it disappeared for the estrogen-deficient GH-deficient
females vs. controls. Note that WHR was greater in the
GH-deficient patients whether they were estrogen deficient or not. BMI
was increased in the estrogen-sufficient patients, and it approached
significance in estrogen-deficient patients. There was no difference in
IMT between patients and controls regardless of estrogen status (Table 4).
|
). Because of the relatively small
sample size, patients and controls were combined for this analysis. As
shown in Table 5, GH deficiency predicted all variables with the
exception of E-selectin when all other variables were controlled for.
BMI or WHR, i.e. degree of body fat, was independently
predictive of E-selectin, vWF, and EDD. Moreover, in females who had
the greater endothelial dysfunction, GH deficiency was predictive of
all parameters except E-selectin; BMI was independently predictive of
E-selectin and vWF but not EDD. In males, there were few predictive
variables, which is not surprising given their more modest changes in
parameters of endothelial dysfunction. However, WHR and not GH
deficiency was predictive when a relationship was demonstrated. As
expected from previous studies, IMT was predicted by age and blood
pressure in both sexes.
|
We have shown that patients with hypopituitarism and severe GH deficiency display significant abnormalities of several biochemical markers of endothelial cell activation and that these are accompanied by impaired endothelium-dependent brachial artery dilatory responses to occlusion ischemia. Together, these markers imply the presence of early atherosclerosis in such patients and suggest that endothelial dysfunction may contribute to the reduced life expectancy resulting from excess vascular events in these hypopituitary patients. Moreover, the negative correlation of these biochemical markers with IGF-I suggests that GH deficiency may play a role in the genesis of these abnormalities. This was further supported by the stepwise linear regression analysis, in which GH deficiency was a determinant of EDD and biochemical markers, particularly in women. We could not demonstrate any difference in IMT between patients and controls in this study, in contrast to the earlier reports showing differences at baseline between patients and controls (6, 7, 8), which was improved after treatment with GH therapy (8). It is possible that our patient group is different; the patients in the studies of Markussis et al. (6) and Capaldo et al. (7) were relatively younger, and Pfeifer et al. (8) studied only young males.
To date, only a few studies have reported evidence of endothelial dysfunction in hypopituitarism (8, 23, 24, 25). Boger et al. (23) examined urinary nitrite levels and cyclic GMP as markers for endogenous nitric oxide production in patients with hypopituitarism and GH deficiency before and after rhGH therapy. Levels of urinary nitrite and cyclic GMP increased to levels comparable to those of the control group after 1 yr of GH therapy. Evans et al. (24), using a technique similar to ours, reported significant reduction of flow-mediated EDD in 17 GH-deficient adults. Our findings are in agreement with theirs, although we used a larger number of patients. Pfeifer et al. (8) studied brachial artery EDD and carotid artery morphology by high-resolution ultrasonography in 11 patients with GH deficiency before and after GH therapy. That study confirmed normalization of both IMT and EDD after 3 months of therapy, although there was no significant difference in EDD at baseline between patients and controls. More recently, Kvaniscka et al. (25) reported reduction of increased levels of soluble P-selectin, E-selectin, and ICAM-1 as well as other inflammatory markers, to levels comparable to those in the control group, after 1 yr of GH therapy. However, none of these studies examined whether there is any relationship between the biochemical and biophysical indices of endothelial dysfunction. There would be a practical advantage in having a reliable serum marker(s) that correlated with functional changes in endothelial function for following response to treatment.
Cell adhesion molecules mediate the adhesion and migration of monocytes and leukocytes to and across the endothelium, which is a crucial early step in the atherogenic process (15, 29). Soluble forms of ICAM-1 have been shown to predict the risk of future myocardial events in apparently healthy men (30). We have previously reported that levels of E-selectin and ICAM-1 are abnormal well in advance of clinical vascular disease in both adult and young patients with diabetes mellitus (31, 32). Our study group was free from any clinical vascular disease, and occult clinical vascular disease was excluded by assessment of ankle brachial pressure index and baseline electrocardiography. Rhode et al. (33) demonstrated, in a heterogenous group of middle-aged patients referred with symptoms of cardiac disease, a significant correlation of adhesion molecules with carotid artery IMT. Their patients had a high prevalence of dyslipidemia and hypertension (>50%), and almost one third had had previous vascular events. None of our study subjects had significant risk factors for vascular disease, which may explain why we could not demonstrate a relationship between adhesion molecules and IMT.
vWF is found in the Wieble-Palade bodies of endothelial cells and in platelets and is regarded as a more conventional marker of endothelial cell activation in various vasculopathies, including ischemic heart disease (34) and stroke (35). On the other hand, soluble TM was recently recognized to be a new and selective marker of endothelial cell activation in atherosclerotic vascular disease (36). We reported previously that TM might be a marker of endotheliopathy well in advance of clinical vascular disease in young patients with type 1 diabetes (37). The significant correlation of both vWF and TM with E-selectin suggests that the same conclusion applies to hypopituitary GH-deficient patients.
The mechanism(s) responsible for the endothelial dysfunction in our study group is not clear. However, because our patients are well replaced and stable on conventional replacement therapy except GH, GH deficiency may be a contributor. Treatment with GH was shown to improve markers of hemostasis and vascular integrity (38), lipid abnormalities in some but not all of the studies, visceral adiposity, and insulin resistance [reviewed by Caroll et al. (39)]. In addition to the early study by Boger et al. (23), the effect of GH therapy on endothelial cell function has recently been suggested by Serri et al. (40), who demonstrated improvement in cytokine production and monocyte adhesion after GH treatment in GH-deficient adults. Expression of cell adhesion molecules by the endothelium is cytokine mediated (17). Several groups have reported that short-term rhGH treatment improves endothelial abnormalities in GH-deficient adults (8, 22, 41, 42). The negative correlation of ICAM-1 and some biochemical indices of endothelial dysfunction with IGF-I in this study suggests an involvement of the GH/IGF-I axis in these mechanisms. Receptors for IGF-I have been characterized in the vascular endothelium (43), and the hemodynamic effect of IGF-I is known to be mediated by nitric oxide (44). The generation of nitric oxide from endothelial cells has been shown to be IGF-I dependent (45). Therefore, it can be argued that IGF-I deficiency may be an initiating mechanism of the early atherosclerosis in GH deficiency.
Increased BMI, WHR, and fat mass are well-recognized features of GH deficiency (9), and these themselves are predictors of vascular disease. Thus, we sought to determine whether they were predictors of endothelial dysfunction independent of GH deficiency. This was certainly true for men but less so for women, who had more marked abnormalities, and for the cohort as a whole. Even where BMI was a predictor in women and the whole cohort, this was independent of GH deficiency with the exception of E-selectin, which was only predicted by BMI. Thus, GH deficiency itself is an important determinant of these early markers of vascular disease, irrespective of its relationship with fat mass.
Abnormalities of lipoprotein metabolism also may be an important mechanism of endothelial cell injury. This may be mediated via oxidative stress, because lipid peroxidation is a rich source of free radicals. GH-deficient patients are known to have significant abnormalities of lipoprotein metabolism (46, 47), so in theory, lipid peroxidation may be enhanced in these patients. Increased free radical production and oxidative stress in GH deficiency has been suggested in a preliminary report (48) that showed excess malondialdehyde levels in a group of GH-deficient adults. The same group recently showed that GH treatment may improve oxidative stress in these patients (41). Such findings were not confirmed by Serri et al. (40). However, both studies involved small numbers of patients, so further studies are required to explore these relationships.
Other mechanisms of endothelial dysfunction in our study group may be underreplacement or overreplacement with conventional hormones. Excess glucocorticoids is known to be associated with visceral adiposity, lipid abnormalities, hyperinsulinemia, and glucose intolerance (49, 50). The current available method of replacing glucocorticoids is not physiological, and concern for undue tissue exposure to excess glucocorticoids during the day with underreplacement at night has been expressed (51, 52). The possible contribution of such effects to the observed excess vascular risk in GH-deficient adults was highlighted recently (53, 54).
Another important factor may be the adequacy of sex hormone replacement in hypopituitary patients, because this seems to contribute to their excess vascular mortality (2). Sex hormones play a crucial role in the cardiovascular system and vascular risk (55, 56). The female patient group (as a whole) had more abnormalities of both the biophysical and biochemical markers of endothelial dysfunction than their male counterparts. With regard to EDD, the difference was more striking in the estrogen-deficient female subgroup. This suggests a significant contribution of sex hormone deficiency to the biophysical abnormalities observed in this study. Hormone replacement therapy in postmenopausal women is known to improve impaired vasodilatory response (57, 58), and hormonal changes during the normal menstrual cycle in young women influence endothelium-dependent vasodilation (59, 60). This protective effect on endothelial function is thought to be mediated by enhanced nitric oxide release, possibly through stimulating nitric oxide synthase (61), just as estradiol has been shown to trigger nitric oxide release in animals (62). Therefore, it is plausible that the excess mortality seen in hypopituitary GH-deficient adults, especially in women in the recent observation by Tomlinson et al. (2), is related to early abnormalities in endothelial cell function in these patients, as shown by this study.
In contrast, the effect of GH deficiency is more striking for the biochemical markers of endothelial cell activation, because the difference was more marked in the estrogen-replete women. It is possible that GH deficiency has more influence on the biochemical component of endothelial cell function, especially taking into account the negative correlation of these markers with IGF-I, and that these markers are normalized in GH-deficient hypopituitary adults after rhGH therapy (25). The lack of difference in the male groups supports this idea, because all but one of the male GH-deficient patients who were androgen deficient were receiving adequate testosterone replacement.
In conclusion, we have demonstrated significant relationships between abnormalities of biochemical and biophysical markers of endothelial dysfunction in hypopituitary GH-deficient adults, and we have shown that these markers are amplified by the combination of GH and gonadal hormone deficiencies. This may provide new insights into our understanding of the mechanism by which vascular events contribute to the reduced life expectancy of patients with adult-onset hypopituitarism. Furthermore, it will be interesting to study the effects of GH treatment in these patients, because such markers may help to monitor the response of such patients to rhGH therapy.
Acknowledgments
We thank to Dr. Sheila Bone for invaluable support for this project. We also thank Dave Olley (Department of Nuclear Medicine, North Staffordshire Hospital NHS Trust) for performing the biophysical measurements.
Footnotes
This study was supported by a generous educational grant from Pharmacia & Upjohn, Inc.
Abbreviations: BMI, Body mass index; EDD, endothelium-dependent vasodilation; ICAM-1, intercellular adhesion molecule-1; IMT, intima-media thickness; TM, thrombomodulin; vWF, von Willebrand factor; WHR, waist to hip ratio.
Received January 5, 2001.
Accepted May 3, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  A. Colao, C. Di Somma, S. Spiezia, S. Savastano, F. Rota, M. C. Savanelli, and G. Lombardi Growth Hormone Treatment on Atherosclerosis: Results of a 5-Year Open, Prospective, Controlled Study in Male Patients with Severe Growth Hormone Deficiency J. Clin. Endocrinol. Metab., September 1, 2008; 93(9): 3416 - 3424. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A van der Klaauw, A. M Pereira, T. J Rabelink, E. P M Corssmit, A.-J. Zonneveld, H. Pijl, H. C de Boer, J. W A Smit, J. A Romijn, and E. J P de Koning Recombinant human GH replacement increases CD34+ cells and improves endothelial function in adults with GH deficiency Eur. J. Endocrinol., August 1, 2008; 159(2): 105 - 111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Sen, M. Demirturk, N. Abaci, E. Golcuk, H. Oflaz, A. Elitok, F. Kutluturk, H. Issever, N. E. Unaltuna, and N. C. Ozbey Endothelial nitric oxide synthase intron 4a/b polymorphism and early atherosclerotic changes in hypopituitary GH-deficient adult patients Eur. J. Endocrinol., May 1, 2008; 158(5): 615 - 622. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Ungvari, C. Parrado-Fernandez, A. Csiszar, and R. de Cabo Mechanisms Underlying Caloric Restriction and Lifespan Regulation: Implications for Vascular Aging Circ. Res., March 14, 2008; 102(5): 519 - 528. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Devin, L. S. Blevins Jr., D. K. Verity, Q. Chen, J. R. Bloodworth Jr., J. Covington, and D. E. Vaughan Markedly Impaired Fibrinolytic Balance Contributes to Cardiovascular Risk in Adults with Growth Hormone Deficiency J. Clin. Endocrinol. Metab., September 1, 2007; 92(9): 3633 - 3639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Feldt-Rasmussen, M. Lange, W. Sulowicz, U. Gafter, K. N. Lai, J. Wiedemann, J. S. Christiansen, M. El Nahas, and and the Adult Patients in Chronic Dialysis (APCD) Growth Hormone Treatment during Hemodialysis in a Randomized Trial Improves Nutrition, Quality of Life, and Cardiovascular Risk J. Am. Soc. Nephrol., July 1, 2007; 18(7): 2161 - 2171. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bollerslev, T. Ueland, A. P Jorgensen, K. J Fougner, R. Wergeland, T. Schreiner, and P. Burman Positive effects of a physiological dose of GH on markers of atherogenesis: a placebo-controlled study in patients with adult-onset GH deficiency. Eur. J. Endocrinol., April 1, 2006; 154(4): 537 - 543. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Lanes, A. Soros, K. Flores, P. Gunczler, E. Carrillo, and J. Bandel Endothelial Function, Carotid Artery Intima-Media Thickness, Epicardial Adipose Tissue, and Left Ventricular Mass and Function in Growth Hormone-Deficient Adolescents: Apparent Effects of Growth Hormone Treatment on These Parameters J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 3978 - 3982. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, C. Di Somma, F. Rota, R. Pivonello, M. C. Savanelli, S. Spiezia, and G. Lombardi Short-Term Effects of Growth Hormone (GH) Treatment or Deprivation on Cardiovascular Risk Parameters and Intima-Media Thickness at Carotid Arteries in Patients with Severe GH Deficiency J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2056 - 2062. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Reis, M. V. Campos, M. Bastos, L. Almeida, M. Lourenco, C. A. Ferrer-Antunes, A. Palmeiro, J. D. Santos-Dias, J. F. Mesquita, M. Carvalheiro, et al. Platelet Hyperactivation in Maintained Growth Hormone-Deficient Childhood Patients after Therapy Withdrawal as a Putative Earlier Marker of Increased Cardiovascular Risk J. Clin. Endocrinol. Metab., January 1, 2005; 90(1): 98 - 105. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. K. Hansen, S. Fisker, R. Dall, T. Ledet, J. O. L. Jorgensen, and L. M. Rasmussen Growth Hormone Increases Vascular Cell Adhesion Molecule 1 Expression: in Vivo and in Vitro Evidence J. Clin. Endocrinol. Metab., February 1, 2004; 89(2): 909 - 916. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Serri, L. Li, F. Maingrette, N. Jaffry, and G. Renier Enhanced Lipoprotein Lipase Secretion and Foam Cell Formation by Macrophages of Patients with Growth Hormone Deficiency: Possible Contribution to Increased Risk of Atherogenesis? J. Clin. Endocrinol. Metab., February 1, 2004; 89(2): 979 - 985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Smith, H. A. Lane, J. Lewis, S. Dann, J. Goodfellow, P. Collins, L. M. Evans, M. F. Scanlon, and J. S. Davies Endothelial Function and Coagulant Factors in Growth Hormone-Treated Hypopituitary Adults Receiving Desmopressin J. Clin. Endocrinol. Metab., May 1, 2003; 88(5): 2152 - 2156. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Elhadd, T. A.M. Abdu, and R. N. Clayton Excess Vascular Mortality in Hypopituitarism: Is it the Result of Estrogen or GH Deficiency? J. Clin. Endocrinol. Metab., July 1, 2002; 87(7): 3509 - 3509. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||
