Varicose veins of the lower limbs and venouscapacitance in postmenopausal women:Relationship with obesity
Arcangelo Iannuzzi, MD,a Salvatore Panico, MD,b Anna V. Ciardullo, MD,d Cristina Bellati, MD,e
Vincenzo Cioffi, MD,b Gabriella Iannuzzo, MD,b Egidio Celentano, MD,c Franco Berrino, MD,e
andPaolo Rubba, MD,b Salerno, Naples, and Milan, Italy
Objective: The purpose of this study was to examine the association between body mass index (BMI), venous capacitance,
and clinical evidence of varicose veins after adjustment for sex hormones in postmenopausal women.
Methods: This study group of the DIANA (DIet and ANdrogens) project (a randomized controlled trial on the effect of
some dietary changes on sex hormone pattern in women with elevated androgenic hormone levels in Italy) was comprised
of 104 healthy volunteer postmenopausal women, aged 48 to 65 years. The main outcome measures were physical
examination to determine the presence and severity of varicose veins and plethysmographic measurement of lower limb
venous capacitance and outflow.
Results: Women in the upper quartile of BMI (>30 kg/m2) showed a positive association with clinical evidence of varicose
veins (odds ration, 5.8; 95% CI, 1.2 to 28.2) after adjustment for age, estradiol, testosterone, and sex hormone binding
globulin. No association was found between BMI and plethysmographic measurements of venous parameters.
Conclusion: Obesity is associated with clinical evidence of varicose veins independently from the influence of sex hormones
in postmenopausal women and is not associated with venous capacitance. Increased body weight increases the risk of
varicose veins. (J Vasc Surg 2002;36:965-8.)
Venous disease of the lower limbs is a major problem
veins, and plethysmographic measurement of venous ca-
affecting Western societies that results in considerable mor-
pacitance, after controlling for age and sex hormone pat-
bidity in the population and cost to the healthcare service.1
tern (ie, serum estradiol, sex hormone binding globulin
Many epidemiologic studies on the prevalence of varicose
[SHBG], and testosterone), in a group of healthy post-
veins have found an association between body mass index
menopausal women to clarify the role of obesity in the
(BMI) and presence of varicose veins.2-4 Other factors
pathogenesis of venous dysfunction in the lower limbs.
could play a role in the development of varicose veins,particularly sex hormone patterns.5 To our knowledge, no
PARTICIPANTS AND METHODS
study has investigated the association between BMI, vari-
Progetto DIANA (DIet and ANdrogens) is
cose veins, and lower limb venous compliance, taking into
a nutritional trial in the Milan area, Northern Italy, within
account the concentration of sexual hormones of the indi-
the research activities of the Epidemiology Unit of the
National Cancer Institute. The general objective of the
One of the factors associated with primary venous
project was to test the effect of a specific diet on the levels of
dysfunction is an abnormal venous wall distensibility.6 Es-
serum testosterone together with other effects on sex and
trogens promote the appearance of varicose veins because
metabolic hormones in healthy female volunteers aged 48
of an increased venous capacitance.5 Strain-gauge plethys-
to 65 years invited through newspapers and broadcasting
mography is one of the best available methods to study
advertisements.8 Potential participants had to fulfill the
maximal venous capacitance.7 In this study, we evaluated
following conditions: postmenopausal for at least 2 years,
the relationships among BMI, clinical evidence of varicose
no history of ovariectomy, no hormonal replacement treat-ment for at least 6 months, no history of cancer or vasculardisease (in particular no previous venous thromboembo-
From the Division of Internal Medicine, Cava de’ Tirreni and Amalphitan
lism), no adherence to vegetarian or macrobiotic diet or to
Coast Hospitala; the Department of Clinical and Experimental Medicine,Federico II Universityb; the Epidemiology Unit, National Cancer Insti-
any other diet prescribed for medical reasons, no treatment
tutec; CeVeas, ASL Modenad; and the Division of Epidemiology, Na-
for diabetes, and agreement of participation in the study
Three hundred and twelve women were recruited.
Supported by research grants from Fondazione CARIPLO and Europe
However, because one of the main objectives of this project
Reprint requests: Arcangelo Iannuzzi, MD, Via F. Terracciano, 87, Pomigli-
was to reduce testosterone levels through a nutritional
ano d’Arco, 80038, Napoli, Italy (e-mail: [email protected]
approach, 104 were selected to participate in the trial on the
Copyright 2002 by The Society for Vascular Surgery and The American
basis of their serum testosterone levels (included in the
upper tertile). An ancillary study of this nutritional project
0741-5214/2002/$35.00 ϩ 0 24/1/128315
examined the relationship of varicose vein prevalence and
966 Iannuzzi et al
lower limb venous function with BMI in this group of
two cuffs measuring 22 ϫ 70 cm were wrapped around the
thighs and two tension detectors (strain-gauge), with un-
Written informed consent to participate in the study for
stretched length about 90% of the circumference of the
research purposes was obtained from all women before
limb, were fixed around the calves at the point of maximum
enrollment. The Scientific and Ethical Committee of the
circumference. The cuffs began to inflate at 50 mm Hg, and
National Cancer Institute, Milan, Italy, approved the re-
the compression was kept until a relatively stable calf vol-
ume was achieved. The incremental volume represents a
Baseline observation of all the women re-
measure of the quantity of blood the examined district is
cruited in the study included a physical examination for
able to receive because of its distension and is called venous
diagnosis and classification of varicose veins, a venous
capacitance or maximal incremental venous volume
strain-gauge plethysmographic measurement of venous ca-
(MVIV). MVIV is expressed in mL/100 mL volume.
pacitance and outflow, standard anthropometric measure-
Limbs with varicose veins have the largest venous capaci-
tance.7 After that, the cuffs were rapidly deflated. The
Blood samples were collected
downward slope is the function of emptying speed of the
from the patients in the morning after an overnight fast.
previous venous pooling of the leg and is measured as the
Serum for hormonal assays was stored at Ϫ30° C for a short
maximum venous outflow and expressed in mL/100 mL/
time and then at Ϫ80° C. Circulating hormones were
measured with commercial kits: RIA kits were purchased
Five women were not present at
from ORION Diagnostic (Turku, Finland) for testosterone
the venous vessel examination. For two patients, the results
and estradiol (tailored for postmenopausal condition) and
of the sex hormone determination were not available.
IRMA kits from Farmos (Oulunsalo, Finland) were pur-
Therefore, the final comparisons were based on 97 women.
We tested the association between both the dichotomous
Clinical examination of legs.
Two medical research
variable varices (yes/no) and the MVIV (upper tertile
fellows trained in internal medicine with qualified experi-
versus others) with BMI. Adjustment for age, estradiol,
ence in angiology together examined all the women. At the
testosterone, and SHBG values as covariates was performed
first inspection, with the patient erect, they judged the
in the model because of the relationship of estradiol with
varicose veins as trunk varices (dilated trunks of the long or
varices shown in this study in the logistic regression analysis
short saphenous veins or their principal branches), reticular
(highest tertile versus others: OR, 3.0; 95% CI, 1.1 to 8.8)
varices (dilated or tortuous superficial veins that did not
and the correlation of SHBG and testosterone with BMI
belong to the main trunk or its major branches), or hyphen-
shown in the Pearson correlation matrix (Ϫ0.40; P
web varices (intradermal varices). During the clinical exam-
Ͻ .05, respectively). We computed odds ratios and
ination, the varicose veins were marked with the patient
95% CIs with logistic regression analysis with SPSS Win-
erect and then with the patient supine and the leg elevated.
dows 98 release 10.0 (SPSS, Inc, Chicago, Ill).
The fascial defects, which were often tender, were markedas possible perforating veins. Subsequently the saphe-
nofemoral junction and the previously marked defects were
Descriptive features of the women with trunk varices
controlled by the fingertips and the women were asked to
(cases) and control subjects are reported in Table I. Of 97
stand. Control was tested with removal of the fingers in
women, 31 (32%) had clinical varices (C ϩ C ) and 66
turn, examining for venous reflux at the fascial defects. This
(68%) had no visible or palpable signs of venous disease or
method of examination has been indicated to be more
only telangiectases or reticular veins (C ϩ C ).
effective than the traditional Trendelenburg’s test.9 Subse-
After the CEAP classification, 55 women had C (no
quently, the CEAP classification10 was used to discriminate
visible or palpable signs of venous disease), 11 had C1
the chronic venous diseases of the lower limbs on the basis
(telangiectases or reticular veins), 29 had C
of clinical signs, etiology, anatomic involvement of the
veins), and two had C (edema). Other authors have shown
superficial, perforating, or deep veins, and pathophysiology
that only trunk varices are associated in women with leg
leading to evidence. Duplex scanning of the veins of the
symptoms such as heaviness or tension, aching, and itch-
lower limbs was not performed to exclude deep venous
ing,11 so we considered in our data analyses trunk varices
thrombosis. However, all the participants in the study
(31 patients) as cases and we added in the control group the
underwent a plethysmographic study, and all had normal
patients with only telangiectases or reticular veins (11 pa-
values of maximum venous outflow, which is a sensitive
tients). No skin changes were ascribed to venous disease in
the women participating the study. Ten of 31 women with
On the same morning of the clin-
varicose veins had telangiectases or reticular veins, too. All
ical examination, all the women underwent a plethysmo-
the women had primary varicose veins. Twenty-four of the
graphic examination of lower limb venous capacitance and
women with trunk varices had venous disease only in the
outflow. The methodology has been described in detail in a
superficial veins, and seven had venous disease in the super-
previous paper.5 However, a brief description of the prin-
ficial and perforating veins. The clinical signs and symptoms
cipal points follows. With the patient on an examination
of venous dysfunction in all the cases were the result of
bed with legs slightly elevated to facilitate venous drainage,
JOURNAL OF VASCULAR SURGERYVolume 36, Number 5
Iannuzzi et al 967
Characteristics of study participants (97 women)
Odds ratios (95% CI) of varices and MVIV in
relation to BMI
*Adjusted for age, estradiol, sex hormone binding globulin, and testoster-
in a setting that allows the availability of serum concentra-
tion of sex hormones and plethysmographic measurements
of venous capacitance in postmenopausal women. Theprevalence of varicose veins varies among different popula-
All values are mean Ϯ standard error of mean.SBP,
Systolic blood pressure; DBP,
diastolic blood pressure; HDL
tions. Its prevalence is low in African or Australian aborig-
density lipoprotein; MVO,
maximum venous outflow.
ine populations (0 to 5%)12,13 and is high in Westerncountries (25% to 75%).14-17 Different epidemiologic ter-
The coefficients of intraassay and interassay variations
minology, population sampling, and varicose vein defini-
of eight replicates for each hormone analyzed were: 4.2%
tions account for much of the variation in the literature. In
and 12.5% for a testosterone value of 0.420 ng/mL; 5.2%
this study, we used the CEAP classification of varicose veins
and 11.1% for an estradiol value of 10 pg/mL; and 3.5%
of the lower limbs to permit a comparison of this survey
and 6.7% for an SHBG value of 34.0 nmol/L.5 Quartiles of
BMI were the following: first quartile, 20 to 23 kg/m2;
In an Italian observation, the prevalence of varicose
second quartile, 24 to 26 kg/m2; third quartile, 27 to 29
veins in the elderly population of the Campania region
kg/m2; and fourth quartile, 30 to 42 kg/m2. The women
(1319 subjects; mean age, 74 years) was 35.2% in wom-
in the upper quartile of frequency distribution had more
en,18 and in agreement with these data, our study reports a
than 30 kg/m2 and were classified as obese. Upper quartile
prevalence of varicose veins in 32% of postmenopausal
versus others of BMI was used as independent variable,
women. In our data, although women were selected for
with varices and MVIV as dependent variables. The women
high testosterone levels, they had the classical characteris-
with BMI in the upper quartile of the frequency distribu-
tics of a large number of postmenopausal women, such as a
tion had a significant association with clinical evidence of
high prevalence of obesity and lower limb varicosity.
varicose veins, after adjustment for age and serum concen-
We found that obesity (BMI Ͼ30 kg/m2) in post-
tration of estrogens, testosterone, and SHBG (Table II).
menopausal women was associated with a higher preva-
Test for linear trend across quartiles was 2 of 4.4 (P
lence of varicose veins, whereas no association was found
Without the adjustment for testosterone, the results did
with the plethysmographic measures of venous capacitance.
not change. The highest quartile of BMI was associated
A possible explanation of this finding is that obesity could
with clinical evidence of varices with an odds ratio of 5.8
hamper the normal blood flow exchange between superfi-
(95% CI, 1.2 to 27.8; P
Ͻ .05). No association was found
cial and deep veins of the lower limbs because of aug-
between BMI and MVIV; we tested that association with
mented adipose and fibrous tissue surrounding veins. The
both linear regression (considering BMI and MVIV as
increase of adipose tissue disturbs the cutaneous venous
continuous variables, R2
ϭ .001; P
ϭ .7) and logistic
circulation and damages the drainage veins, provoking
regression (considering MVIV a dichotomous variable and
stasis and subsequently the appearance of varicose veins.19
dividing BMI in quartiles; Table II).
According to our results, there is a linear trend of aug-
With upper quartiles versus others of waist, hip, and
mented prevalence of varicose veins at increasing levels of
thigh circumferences as independent variables and presence
BMI, and for values exceeding 30 kg/m2, there is a signif-
of varicose veins as a dependent variable, without adjust-
icant excess risk of venous disease in menopause. The
ments, a significant association was found for hip circum-
association between higher BMI and varicosity still remains
ferences (odds ratio, 3.8; 95% CI, 1.1 to 13.5) and for thigh
after adjustment for age, estradiol, testosterone, and SHBG
circumferences (odds ratio, 3.6; 95% CI, 1.0 to 12.9). No
association was detected between waist circumference and
Epidemiologic studies have observed that other factors
varicose veins (odds ratio, 1.5; 95% CI, 0.4 to 4.9).
could play a role in the development of venous varicosity ofthe lower limbs, such as pregnancy, long-standing seden-
tary activities, and positive family history.20-22 Most of
To our best knowledge, this is the first article that
these studies have documented an influence of obesity on
examines the association between BMI and varicose veins
the development of varices, at least in the female gender.
968 Iannuzzi et al
This study confirms these previous observations, clarifies
6. Vanhoutte PM, Corcaud S, de Montrion C. Venous disease: from
that the association between BMI and varicose veins is
pathophysiology to quality of life. Angiology 1997;48:559-67.
independent of sex hormone levels, and suggests that other
7. Sumner DS. Strain-gauge plethysmography. In: Bernstein EF, editor.
Non-invasive diagnostic techniques in vascular disease. St Louis–Toron-
physiopathologic mechanisms, different from augmented
to-Princeton: Mosby; 1985. p. 742-54.
venous distensibility, play a role in determining the clinical
8. Berrino F, Bellati C, Secreto G, Camerini E, Pala V, Panico S, et al.
evidence of varicose veins in obesity. In women, the re-
Reducing bioavailable sex hormones through a comprehensive change
gional distribution of fat in obesity is particularly abundant
in diet: the diet and androgens (DIANA) randomized trial. Cancer
in the lower abdomen, gluteal region, and thighs, so it is
Epidemiol Biomarkers Prev 2001;10:25-33.
9. Noble J, Gunn AA. Varicose veins: comparative study of methods for
not surprising that we have found an association between
detecting incompetent perforators. Lancet 1972;1(7763):1253-5.
varicose veins and higher hip and thigh circumferences and
10. Classification and grading of chronic venous disease in the lower limbs.
no association with waist circumferences.
A consensus statement. Ad Hoc Committee, American Venous Forum.
Our results suggest that increased body weight plays a
specific and independent role in the development of vari-
11. Bradbury A, Evans C, Allan P, Lee A, Ruckley CV, Fowkes FGR. What
cose veins of the lower limbs in selected groups of post-
are the symptoms of varicose veins? Edinburgh vein study cross sectionalpopulation survey. Br Med J 1999;318:353-6.
menopausal women with high testosterone levels and give
12. Stanhope JM. Varicose veins in a population of lowland New Guinea.
more support to the concept that obesity is not a simple risk
factor but a specific distinct disease with cardiovascular
13. Richardson JB, Dixon M. Varicose veins in tropical Africa. Lancet
implications. This finding is of clinical interest and provides
another reason to treat overweight patients before a shift to
14. Capitao LM, Menezes JD, Gouveia-Oliveira A. Epidemiological char-
acterization of chronic venous insufficiency in Portugal. Acta Med Port
frank obesity. In most instances, increased body weight is
associated with varicose veins of the lower limbs.
15. Laurikka J, Laara E, Sisto T, Tarkka M, Auvinen O, Hakama M.
Misclassification in a questionnaire survey of varicose veins. J Clin
We thank the Association (Attivecomeprima) for hos-
16. Callam MJ. Epidemiology of varicose veins. Br J Surg 1994;81:167-73.
17. Diehm C. Epidemiology and pathogenesis of varicosities. Herz 1989;
18. Canonico S, Gallo C, Paolisso G, Pacifico F, Signoriello G, Sciaudone
1. Evans CJ, Fowkes FG, Hajivassiliou CA, Harper DR, Ruckley CV.
G, et al. Prevalence of varicose veins in an Italian elderly population.
Epidemiology of varicose veins. A review. Int Angiol 1994;13:263-70.
2. Preziosi P, Galan P, Aissa M, Hercberg S, Boccalon H. Prevalence of
19. Lemaire R. The flow of venous blood in the obese. Phlebologie 1988;
venous insufficiency in French adults of the SUVIMAX cohort. SUpple-
mentation en VItamines et Mineraux AntioXydants. Int Angiol 1999;
20. Stvrtinova V, Kolesar J, Wimmer G. Prevalence of varicose veins of the
lower limbs in the women working at a department store. Int Angiol
3. Sisto T, Reunanen A, Laurikka J, Impivaara O, Heliovaara M, Knekt P,
et al. Prevalence and risk factors of varicose veins in lower extremities:
21. Abramson JH, Hopp C, Epstein LM. The epidemiology of varicose
mini-Finland health survey. Eur J Surg 1995;161:405-14.
veins. A survey in western Jerusalem. J Epidemiol Community Health
4. Lapidus L, Bengtsson C, Hallstrom T, Bjorntorp P. Obesity, adipose
tissue distribution and health in women: results from a population study
in Gothenburg, Sweden. Appetite 1989;13:25-35.
22. Brand FN, Dannenberg AL, Abbott RD, Kannel WB. The epidemiol-
5. Ciardullo AV, Panico S, Bellati C, Rubba P, Rinaldi S, Iannuzzi A, et al.
ogy of varicose veins: The Framingham Study. Am J Prev Med 1988;4:
High endogenous estradiol is associated with increased venous disten-
sibility and clinical evidence of varicose veins in menopausal women. JVasc Surg 2000;32:544-9.
Submitted Mar 21, 2002; accepted Jun 24, 2002.
“I had to win. not One Woman’s Cancer Answer only was my life at an interview by Jeff Gordon www.eatwithoutheat.com stake, but I needed to find help for So just what did you learn eaten the All-American Diet–a diet that, in my about cancer? opinion, that greatly contributes to today’s others as well!” Two major conclusions Great American
Distance Learning (Correspondence) Jewellery Design, Manufacture and Repair: Level: Certificate This entire course is carried out via Distance Learning (correspondence). Distance learning is ideal if you need to adapt around work, family and other commitments. It is also extremely beneficial should you simply wish to fast track your studies and not be re