Morphologic Study of Microcirculation in Acromegaly by ... - CiteSeerX

0021-972X/99/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 1999 by The Endocrine Society

Vol. 84, No. 9 Printed in U.S.A.

Morphologic Study of Microcirculation in Acromegaly by Capillaroscopy* F. SCHIAVON, P. MAFFEI, C. MARTINI, E. DE CARLO, C. FAIS, S. TODESCO, N. SICOLO

AND

Department of Medical and Surgical Sciences, Division of Rheumatology (F.S., C.F., S.T.), Third Medical Clinic (P.M., C.M., E.D.C., N.S.), Padua University, 35100 Padova, Italy ABSTRACT Although wide range investigations on the heart and great vessels have been reported in acromegaly, the field of microcirculation is still largely vacant. The nailfold is a window through which we can observe in vivo the vascular bed. This study investigates through nailfold capillaroscopy the morphology of cutaneous microcirculation in acromegaly in relationship with the usual hormonal parameters of disease activity. Twenty-five acromegalic patients and 26 normal subjects, age and sex matched, were studied. A subgroup of acromegalics (8 patients) was considered in stable remission, and the remaining 17 had active disease. Capillaroscopy was performed in each subject by in vivo computer aided stereomicroscopy (magnification, 3400). The following morphological parameters were calculated: the number of tortuous loops, meandering capillaries, and capillaries per millimeter; avascular areas; visibility of subpapillary plexus; the capillary length; and intercapillary distance. We were unable to perform the exam in 4 of 25 patients because visibility was poor. The capillary number and length were signifi-

A

CROMEGALY is characterized by a premature death rate mainly related to cardiovascular diseases either directly or indirectly (1) caused by the hypersecretion of GH and/or insulin-like growth factor I (IGF-I) (2, 3). Several studies reported peculiar morphological alterations of the heart and great vessels (4, 5), but, surprisingly, neglected evaluation of the microcirculation in acromegaly (3). Conversely, we can assume that acromegaly, similar to other diseases causing cardiovascular alterations such as diabetes and hypertension, may induce morphological changes in the peripheral microcirculation in addition to the alterations of the macrocirculation (6, 7). The importance of microcirculation, even from a clinical point of view, is due to the fact that the capillaries represent a vascular area where trophicmetabolic exchanges between blood and tissue take place, thus leading to a critical role in organ function. Nailfold capillaroscopy is an imaging method allowing the in vivo and atraumatic assessment of the morphology and some functional aspects of cutaneous capillaries. Although it is mainly indicated in the early detection of scleroderma microangiopathy and in connective tissue diseases, it is also used in those diseases whose pathogenesis is due to a anaReceived July 17, 1998. Revision received April 28, 1999. Accepted May 17, 1999. Address all correspondence and requests for reprints to: Dr. Pietro Maffei, Dipartimento di Scienze Mediche e Chirurgiche, Clinica Medica 3, Via Ospedale 105, 35100 Padova, Italy. E-mail: [email protected]. * This work was supported by MURST 40%, 1997 (no. 9706151106).

cantly reduced in acromegalics compared to controls [8.9 6 1.5 vs. 10.3 6 1.2 no./mm (P 5 0.0010) and 174 6 49 vs. 255 6 24 mm (P , 0.0001)]. Moreover, in acromegalics, the numbers of tortuous loops and meandering capillaries were significantly increased [19 6 8 vs. 13 6 5 (P 5 0.0027) and 10 6 12 vs. 0.7 6 1.1 (P , 0.0001)]. The capillaroscopic alterations were still observed in a smaller group of 8 nondiabetic and nonhypertensive acromegalics. We found branch-like capillaries in 4 acromegalic patients, but not in the control group. Finally, we observed a meaningful different and ameliorated capillaroscopic morphology in acromegalic patients in stable remission compared to active disease patients as far as the total number (density) and meandering capillaries were concerned. In conclusion, our study shows that in acromegaly, morphological alterations also affect the peripheral microcirculation, which seems to be influenced by the activity of the disease. We believe that nailfold capillaroscopy may represent an additional useful tool in the follow-up of acromegalic patients. (J Clin Endocrinol Metab 84: 3151– 3155, 1999)

tomical or functional anomaly of microcirculation (8 –13). The noninvasiveness, reproducibility, low cost, and considerable sensitivity make this technique very useful in the early diagnosis and monitoring of microangiopathies. Therefore, this study investigates through nailfold capillaroscopy, from a morphological viewpoint, the condition of cutaneous microcirculation in acromegaly in relationship to the usual hormonal parameters of disease activity. Subjects and Methods Patients The prominent clinical features of affected patients with acromegaly are reported in Tables 1 and 2. A series of 25 acromegalic patients (12 males and 13 females; aged 30 –74 yr) entered the study after informed consent had been obtained. The cases taken into account were studied consecutively. The diagnosis was made according to the clinical picture, basal GH plasma levels (mean of three blood samples), IGF-I, and the paradox response of GH to dynamic tests (oral glucose tolerance test and TRH test). The volume of the pituitary tumor was assessed by means of computed tomography and/or magnetic resonance imaging in all patients. Treatments for hypertension, diabetes, pituitary deficiency, or acromegaly were continued without change. All patients (Table 2) were previously submitted to pituitary surgery, 4 of whom received additional radiotherapy; 5 patients were receiving stable pituitary replacement therapy, which consisted of T4 in 4 patients, 2 of whom required additional cortisone acetate. One female patient received estrogens. A subgroup of patients, without disease activity after surgery and/or radiotherapy and/or medical therapy, was selected. According to these parameters, 8 patients (no. 3, 9, 10, 11, 17, 18, 20, and 25) were considered in stable remission (mean, 9 yr; range, 2–18 yr; IGF-I, 217 6 173 ng/mL;

3151 Downloaded from jcem.endojournals.org by on April 14, 2008

3152 TABLE 1. Clinical characteristics of the 25 patients Characteristics

Acromegalics

Familiar medical history Ischemic heart disease Diabetes mellitus Obesity Lipid disorders Hypertension Endocrine disorders Past medical history Ischemic heart disease Diabetes mellitus Lipid disorders Hypertension Other endocrine disorders Menopause Current smoker Alcohol consumption Coffee consumption a

JCE & M • 1999 Vol 84 • No 9

SCHIAVON ET AL.

18 (72)a 13 (52) 12 (48) 9 (36) 11 (44) 7 (28) 7 (28) 10 (40) 12 (48) 12 (48) 5 (20) 9 (75) 4 (16) 16 (64) 23 (92)

Percentages are shown in parentheses.

GH, 1.9 6 1.7 ng/mL), and the remaining 17 had active disease (IGF-I, 698 6 148 ng/mL; GH, 12.2 6 12.2 ng/mL). From the metabolic point of view the acromegalics were also classified according to the National Diabetes Data Group criteria (14): 15 patients with normal glucose tolerance, 4 patients with impaired fasting glycemia (IFG), 5 patients with impaired glucose tolerance (IGT), and 1 patient with diabetes. A control group of 26 normal subjects, age (acromegalics vs. controls, 52.6 vs. 57.2 yr; P 5 NS) and sex (acromegalics vs. controls, male/female ratio, 12:13 vs. 13:13; P 5 NS) matched, were also studied.

Methods The morphological pattern of the capillaries in the same finger of an individual is surprisingly constant, and its aspect is uniform in healthy subjects regardless of age (13, 15). In most areas of the fingers the nutritional capillary loops are oriented at 90° to skin surface, and only the tip of the capillary loops can be visualized. Usually there are one to three capillaries in each dermal papilla. In the nailfold area, the capillary loops become more parallel to the skin surface, and in the last row they can be seen in their full length. The capillaries include an arterial and a venous limb and the apical part between these two. The arterial portion is generally narrower than the venous one, with an increasing diameter from the proximal arteriolar to the distal venous side. The capillary diameter is represented by the erythrocyte column (13). All acromegalic patients and controls (Fig. 1) were submitted to nailfold capillaroscopy performed by in vivo widefield capillary microscopy using a Leitz stereomicroscope (Leitz Wild Apo 5, Rockleigh, NJ) giving a three-dimension vision. The stereomicroscope is connected to a JVC-TK 1280E color videocamera, which transmits a 400-fold magnified image onto a personal computer (AST-Bravo; Casti Imaging, Venice, Italy). Analogic video signals are digitalized by a real time video digitizer board Matrox Marvel II Frame Grabber. Before investigation the patients were acclimatized in a room with constant temperature for 30 min. The exam is generally carried out on the last four fingers positioned at heart level. A drop of immersion oil is applied to the nailfold to improve the transparency and capillary visibility. Oblique illumination is accomplished by a guided light and a condenser lens that focuses the light on the nailfold. The angle and direction of illumination may be changed to minimize reflections. The fiber optics are coupled to a halogen lamp through a heat absorption filter. Finally, quantitative and semiquantitative assessments of the images are performed by a computer-aided system for morphometric analysis (15). In each case we evaluated the following parameters: 1) the number of tortuous loops (i.e. loops showing at least one criss-crossing of the limbs); 2) the number of meandering capillaries (loops showing three or more criss-crossings); 3) capillary density (loops, number per mm) or the number of capillaries per millimeter in the first row of loops; and 4) avascular areas, defined as distinct areas in the nailfold where two or more capillaries are missing compared to the density of capillaries in the

remaining row (Lee score). Visibility of subpapillary plexus according to a semiquantitative Maricq’s score (16, 17), capillary length (microns), and intercapillary distance (microns) were also calculated. All capillaroscopic evaluations were carried out by a single operator, who was unaware of the hormonal situation, clinical state, and present or past therapy of each acromegalic patient. For obvious reasons, due to the typical somatic features of these patients, it was impossible to carry out a completely blind capillaroscopy study.

Data analysis The statistical analysis and the gathering of data were performed using an Endeavor-4DX2/66L computer through the statistic analysis program STATISTICA (StatSoft, Tulsa, OK). We used the Shapiro-Wilks’ W test as a normality test for the distribution of variables. The analysis between categoric and numeric normal distribution variables was carried out using the t test for nonmatched data, whereas the Mann-Whitney U test was used for the continuous variables. The ANOVA test was used for the comparison among different groups, and the least significant difference test was used for the post-hoc analysis. The x2 test with Yates correction was used to analyze categorical variables. All parameters, when not specified, are expressed as the mean 6 sd. P , 0.05 was considered statistically significant. All data are two-sided.

Results

The capillaroscopic exam in acromegalic patients was difficult to carry out mainly due to the presence of interstitial connective edema, which was responsible for a reduction in the visibility of the capillary bed (flou effect). In fact, we were unable to perform the exam in 4 of 25 patients (no. 9, 12, 21, and 22) because visibility was poor. These 4 patients, however, showed no differences in metabolic, hormonal, or anthropometric parameters with respect to other acromegalic patients. Furthermore, it was impossible to calculate the number of meandering capillaries in 2 patients (no. 7 and 8) and the number of meandering and tortuous capillaries in one (no. 15). In one patient (no. 15) the capillaroscopic exam was carried out only after (unsuccessful) pituitary surgery due to the increased flou effect encountered before operation. No significant correlation was observed between capillaroscopic parameters and total body water, extracellular water, and intracellular water measured by means of bioimpedance (data not shown). The number of capillaries per mm, identifiable in acromegalic subjects, was significantly reduced (8.9 6 1.5 vs. 10.3 6 1.2 no./mm; P 5 0.0010), shorter in length (174 6 49 vs. 255 6 24 mm; P , 0.0001), and with an irregular course (19 6 8 vs. 13 6 5 no.; P 5 0.0027) compared to that in controls (Table 3). No patient, however, presented avascular areas. The intercapillary distance and subpapillary plexus were not significantly different between the two groups (110 6 26 vs. 118 6 22 mm and 0.4 6 0.4 vs. 0.8 6 1.0 mm, respectively). In addition, in acromegalic patients a clear increase in meandering-like (Fig. 2) capillaries was found (10 6 12 vs. 0.7 6 1.1; P , 0.0001). The capillaroscopic morphological alterations observed in the group of acromegalic patients were still observed in a smaller group of 8 nondiabetic and nonhypertensive acromegalic patients (Table 4). Interestingly, we found branch-like capillaries in 4 acromegalic patients (no. 16, 17, 19, and 24), three of whom had disease activity, but not in the control group. No significant correlation was found between capillaroscopic parameters and the age of patients or controls. However, a more notable difference in

Downloaded from jcem.endojournals.org by on April 14, 2008

MICROCIRCULATION IN ACROMEGALY

3153

TABLE 2. Clinical and laboratory findings of the patients Patient no.

Sex

Age (yr)

BMI (kg/m2)

Duration of disease (yr)

Previous treatment

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

M M F M M M F M M F F F F F F F F F M F F M M M M

47 59 58 43 57 63 55 58 72 55 63 30 74 56 35 53 62 50 41 34 50 34 49 56 62

39.3 27.1 29.5 28.4 30.5 29.4 24.8 29 34.8 31.5 24.8 22.9 28.5 26.5 36.3 38.1 24.5 34.9 26.8 25.7 25.6 32.5 31.2 30.5 27.8

8 26 10 14

S1O S S1O S S1O S1O S S1O S S1R S S1O S1O S S1O S1B1O S1R S S1R S1R S S1O S S1B1D1O S

20 26 11 40 12 27 12 34 11 4 15 10 2 16 13 15 12 18 18 20

Replacement therapy

Current treatment

IGF-I (ng/mL)

GH (ng/mL)

Capillaroscopy

O

958 775 246.1 732.4 673.1 837.6 871 663 161.6 25 192 422 451 615.7 647.8 516 114 484 770 55 627 826 824 665 460

33.5 1.1 0.7 44.7 7.8 6.9 9.6 1.8 0.1 0.05 0.4 4.5 3.4 4.6 4.1 1.8 1.9 1.3 8.5 0.05 1.9 30.1 3.5 1.9 1.3

Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes No No Yes Yes Yes

O

O C1T4 T4 O O O C1T4 E O T4

F, Female; M, male; BMI, body mass index; S, surgery; R, radiotherapy; O, octreotide; B, bromocriptine; D, cabergoline; C, cortisone acetate; T4, levothyroxine; E, estrogens; GH, nadir value during OGTT; Yes, capillaroscopy performed; No, capillaroscopy not performed because visibility was poor. TABLE 3. Capillaroscopic parameters in acromegalics and controls

Avascula areas (Lee score) Tortuous loops (no.) Meandering capillaries (no.) PVS (Maricq’s score) Capillaries number (no./mm) Intercapillary distance (mm) Capillary length (mm) a

FIG. 1. Nailfold capillaroscopy. Capillaroscopic pattern of a control subject. Normally shaped capillaries.

the capillaries was observed in acromegalic females compared to female controls (data not shown). Regarding the hormonal disease parameters, IGF-I was significantly correlated only with the total number of capillaries (r 5 20.54; P 5 0.0113), whereas no significant correlation between capillaroscopic parameters and GH levels or disease duration was observed. We then subdivided the acromegalic population into two groups, according to the presence or absence of residual disease activity (Table 5). Such analysis allowed us to highlight a significantly different and ameliorated capillaroscopic morphological picture in acromegalic patients in stable remission compared to that in active disease patients as far as

Acromegalics (21)a

Controls (26)

P

0.08 6 0.2 19 6 8 10 6 12

0.03 6 0.06 13 6 5 0.7 6 1.1

NS 0.0027 ,0.0001

0.4 6 0.4 8.9 6 1.5

0.8 6 1 10.3 6 1.2

110 6 26

118 6 22

NS

174 6 49

255 6 24

,0.0001

NS 0.0010

The number of patients or controls are shown in parentheses.

the total number (density) and meandering capillaries were concerned. Moreover, no difference was observed between the two groups of acromegalic patients with regard to the length of capillaries, the number of tortuous loops, and the intercapillary distance. The capillaroscopic morphological parameters were not significantly different between the hypertensive and normotensive acromegalic patients. Furthermore, no significant correlation between the systolic and diastolic blood pressure values and the capillaroscopic picture was seen. We did not observe a significantly different capillaroscopic morphology of acromegalics affected by IFG, IGT, or diabetes (total no., 8) vs. that of acromegalics with normal glucose tolerance (total no., 13). However, plasma insulin levels showed a significant relationship to capillary length (r 5 20.58; P 5 0.0056).


3154

SCHIAVON ET AL.

FIG. 2. Acromegaly. Remarkable (meandering) of several capillaries. TABLE 4. Capillaroscopic parameters in nondiabetics nonhypertensive acromegalics and controls

Avascula areas (Lee score) Tortuous loops (no.) Meandering capillaries (no.) PVS (Maricq’s score) Capillaries number (no./mm) Intercapillary distance (mm) Capillary length (mm) a

Acromegalics (8)a

Controls (26)

P

0.22 6 0.4 21 6 7 662 0.3 6 0.3 8.5 6 1.9 123 6 30 184 6 57

0.03 6 0.06 13 6 5 0.7 6 1.1 0.8 6 1.0 10.3 6 1.2 118 6 22 255 6 24

0.0331 0.0012 ,0.0001 NS 0.0059 NS ,0.0001

Number of patients or controls are shown in parentheses.

Discussion

Heart and great vessel disease is well known to negatively affect the prognosis of acromegalic patients (2, 3). However, very little has been reported in the literature regarding microcirculation (18, 19). These studies, mainly autoptic, showed an alteration of the small heart vessels, especially in acromegalic patients with hypertension and diabetes (19). Our study demonstrates, for the first time, an in vivo existence of peculiar morphological alterations of the peripheral microcirculaton in acromegaly. From a morphological point of view, what is more evident in acromegalics is the considerable increase in tortuosity of capillary loops and a reduction in the density and length compared to controls. Furthermore, some capillaroscopic anomalies in the group of patients with nonactive disease seem to be reversible. In fact, in these patients we observed a significant increase in the density and a decrease in the number of the most tortuous (meandering) capillaries. Similar morphological pictures have been observed in diseases such as hypertension or diabetes mellitus, which determined microcirculation involvement too (6, 7). This may suggest that hypertension and diabetes, which often represent a complication in acromegaly, may play a pathogenetic role in the alterations described. In fact, previous studies on nontreated nonacromegalic hypertensive patients showed a reduction in the density of capillaries (20) compared with normotensives and a significant relationship between this morphological parameter and the average diastolic blood pressure (7). However, our study does not provide documentary evidence of any relationship

JCE & M • 1999 Vol 84 • No 9

between blood pressure values and capillaroscopic parameters, although 48% of the acromegalic patients had hypertension. Such data may be ascribable to the antihypertensive treatment administered to our patients, which had not been suspended in view of the capillaroscopic investigation. Besides, evaluating the two groups separately, the evidence of a significant and similar reduction of the density of capillaries compared to that in the control groups seems to exclude the role of hypertension in both normotensives and acromegalic hypertensives (data not shown). The literature outlines an increase in the tortuosity of nailfold capillaries in diabetic patients with associated proliferative retinopathy (6). In our cases diabetic retinopathy was not investigated; however, we observed the presence of branch-like capillaries in two patients affected with IFG and IGT. In the eight acromegalic patients with diabetes (IFG, IGT, or diabetes), no significant capillaroscopic morphological alterations compared with nondiabetics were observed. Such discrepancy could be ascribable to the small number of investigated acromegalics currently affected by diabetes (one patient) and to the fact that the other seven patients showed only an IFG or IGT picture. Capillaroscopic pictures similar to those described in our study were observed in the tetralogy of Fallot in relation to hemoglobin desaturation and/or peripheral oxygenation saturation (21). In acromegaly, the observed reduction in the density of capillaries associated with a possible oxyhemoglobin desaturation often due to sleep apnea (22) may induce a condition of peripheral hypoxia too. Therefore, the tortuous loops observed may be seen as a reaction or setting to peripheral hypoxia. Finally, the capillaroscopic picture may depend upon the direct peripheral action of the growth factors GH and/or IGF-I, thus being consistent with the known vascular alterations already observed in acromegaly affecting macrocirculation. The presence of capillaroscopic morphological alterations in acromegalic patients without diabetes or hypertension leads us to assume that acromegaly per se plays a significant role in these capillary abnormalities. In regard to the possible role of IGF-I, it was demonstrated that in diabetic retinopathy, IGF-I is a growth factor in vitro for human retinal endothelial cells and that in vitro mitogenic effects are suppressed by octreotide (23–25). As a support to the concomitant action of GH on the vascular wall, we underline that in rat aorta, after treatment with rat GH, a relative increase in collagen type I and a reduction of elastin occurred (26). According to the authors of this study, such alterations may influence the elasticity and recoiling properties of aorta. Other studies have show that human GH enhances the production of procollagen type I and fibronectin. These findings suggest that a direct action of GH or one mediated by IGF-I on the anatomical structure of the vessel wall exists. The presence of tortuous loops or meandering capillaries in the peripheral microcirculation in acromegaly may therefore be the morphological result of wall damage involving the vessel elasticity property. Both the morphological evidence and the theoretical and experimental assumptions suggest that the specific capillaroscopic picture is due to acromegaly itself and not concomitant complications, as previously demonstrated in acromegalic cardiomyopathy.


MICROCIRCULATION IN ACROMEGALY

3155

TABLE 5. Capillaroscopic parameters in acromegalics (according to disease remission or activity) and controls

Avascula areas (Lee score) Tortuous loops (no.) Meandering capillaries (no.) PVS (Maricq’s score) Capillaries number (no./mm) Intercapillary distance (mm) Capillary length (mm)

Active acromegalics (14)a

Remission acromegalics (7)

Controls (26)

0.2 6 0.3 17.7 6 8.9 13.6 6 15.0 0.5 6 0.4 8.4 6 1.4 114 6 28 185 6 53

0.1 6 0.1 22.7 6 7.2 5.4 6 2.3d 0.2 6 0.3 9.8 6 1.3f 103 6 20 152 6 29

0.0 6 0.1 13.3 6 4.8b,c 0.7 6 1.1e 0.8 6 1.0 10.3 6 1.2e 118 6 22 255 6 24e,g

a

The number of patients or controls are shown in parentheses. P 5 0.048 controls vs. active acromegalics. P 5 0.001 controls vs. remission acromegalics. d P 5 0.030 remission vs. active acromegalics. e P , 0.001 controls vs. active acromegalics. f P 5 0.017 remission vs. active acromegalics. g P , 0.001 controls vs. remission acromegalics. b c

Although the flou effect, due to the presence of interstitial connective edema, was responsible for a reduction in the visibility of the capillary bed, we do not believe that this may contribute to a relevant bias to our findings. In fact, the connective edema is unable to alter the morphology of the capillaries. Moreover, during the performance of this exam only those capillaries with clearly evident morphological alterations were considered meandering. For this reason a possible bias in the calculation may have only underestimated our results. In summary, our study shows that in acromegaly, morphological alterations also affect the peripheral microcirculation, thus suggesting extended vascular distress. These morphological alterations are compatible with hypoperfusion (capillary number reduction), with possible secondary implications of the trophic function. The clinical repercussions of the morphological alterations could be high if they are present even in tissue considered to be more critical than the epidermis i.e. the myocardium. Finally, the treatment of the disease seems to improve such alterations at least in part. However, the persistence of morphological vascular damage of microcirculation in patients considered to be in disease remission from a hormonal point of view leads to further questions on the long term prognosis of these patients. We believe that the nailfold capillaroscopy, thanks to its simplicity, noninvasiveness, and low cost, may represent an additional useful diagnostic and prognostic tool in the follow-up of acromegalic patients. Acknowledgment We are grateful to Ms. Denise Kilmartin for language revision.

References 1. Maffei P, Plebani M, Sicolo N. 1999 Lipoprotein(a) in acromegaly. Ann Intern Med. 16:537–538. 2. Melmed S. 1990 Acromegaly. N Engl J Med. 322:966 –977. 3. Sacca´ L, Cittadini A, Fazio S. 1994 Growth hormone and the heart. Endocr Rev. 15:555–573. 4. Lusiani L, Ronsisvalle G, Visona` A, et al. 1988 Acromegalic cardiomyopathy. An echocardiographic study. J Endocrinol Invest. 11:159 –164. 5. Sicolo N, Bui F, Sicolo M, et al. 1993 Acromegalic cardiopathy: a left ventricular scintigraphic study. J Endocrinol Invest. 16:123–127.

6. Chang C-H, Tsai R-K, Wu W-C, Kuo S-L, Yu H-S. 1997 Use of dynamic capillaroscopy for studying cutaneous microcirculation in patients with diabetes mellitus. Microvasc Res. 53:121–127. 7. Gasser P, Buhler FR. 1992 Nailfold microcirculation in normotensive and essential hypertensive subjects, as assessed by video-microscopy. J Hypertens. 10:83– 86. 8. Moneta G, Vollenwieder U, Dubler B, Bollinger A. 1986 Diagnostic value of capillaroscopy with and without fluorescent dyes to detect early connective tissue disease. VASA. 15:143–149. 9. Granier F, Vayssairat M, Priollet P, Housset E. 1986 Nailfold capillary microscopy in mixed connective tissue diseases. Comparison with systemic sclerosis and systemic lupus erythematousus. Arthritis Rheum. 29:189 –195. 10. Lefford F, Edwards JCW. 1986 Nailfold capillary microscopy in connective tissue disease: a quantitative morphological analysis. Ann Rheum Dis. 45:741–749. 11. Caspary L, Schmees C, Schoetensack I, et al. 1991 Alterations of tha nailfold capillary morphology associated with Raynaud phenomenon in patients with systemic lupus erythematosus. J Rheumatol. 18:559 –566. 12. Kabasakal Y, Elvins DM, McHugh NJ. 1996 Quantitative nailfold capillaroscopy findings in a population with connective tissue disease and in normal healthy controls. Ann Rheum Dis. 55:507–512. 13. Bollinger A, Fagrell B. 1990 Clinical capillaroscopy. A guide to its use in clinical research and practice. Toronto: Hogrefe & Huber; vii–ix. 14. American Diabetes Association. 1997 Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 20:1183–1197. 15. Bollinger A, Fagrell B. 1990 Clinical capillaroscopy. A guide to its use in clinical research and practice. Toronto: Hogrefe & Huber; 1– 8. 16. Maricq HR. 1970 “Wide-field” photography of nailfold capillary bed and a scale of plexus visualization score. Microvasc Res. 2:335–340. 17. Maricq HR, Le Roy EC, D’Angelo WA, et al. 1980 Diagnostic potential of in vivo capillary microscopy in cleroderma and related disorders. Arthritis Rheum. 23:183–189. 18. Bach R, Leicht E, Langer H-J, et al. 1992 Kardiale funktion und kutane mikrozirkulation bei akromegalie. Dtsch Med Wochenschr. 117:483– 489. 19. Lie JT, Grossman SJ. 1980 Pathology of the heart in acromegaly: anatomic findings in 27 autopsied patients. Am Heart J. 100:41–52. 20. William SA, Tooke J, MacGregor GA. 1986 Rarefaction of skin capillaries in hypertension [Abstract]. Clin Sci. 70(Suppl 13):14. 21. Chang C-H, Yu H-S, Chen G-S, Wu J-R, Huang T-Y, Yu C-L. 1996 Deterioration of cutaneous microcirculation status and its clinical correlation in tetralogy of Fallot. Microvasc Res. 51:59 – 68. 22. Metzon BJ, West P, MacClean J, Kryger M. 1980 Sleep apnea in acromegaly. Am J Med. 69:615– 618. 23. Woltering EA, Barrie R, O’Dorisio TM, et al. 1991 Somatostatin analogues inhibit angiogenesis in the chick chorioallantoic membrane. J Surg Res. 50:245–251. 24. Grant MB, Caballero S, Millard WJ. 1993 Inhibition of IGF-I and b-FGF stimulated growth of human retinal endothelial cells by the somatostatin analogue, octreotide: a potential treatment for ocular neovascularization. Regul Pept. 48:267–278. 25. Danesi R, Del Tacca A. 1996 The effects of the somatostatin analog octreotide on angiogenesis in vitro. Metabolism. 45(Suppl 1):49 –50. 26. Bruel A, Oxlund H. 1995 Biosynthetic growth hormone increases the collagen deposition rate in rat aorta and heart. Eur J Endocrinol. 132:195–199.
