71

Lymphology 45 (2012) 71-79

LIPEDEMA IS ASSOCIATED WITH INCREASED AORTIC STIFFNESS

G. Szolnoky, A. Nemes, H. Gavallér, T. Forster, L. Kemény

Department of and Allergology (GS,LK) and 2nd Department of and Center (AN,HG,TF), University of Szeged; Dermatological Research Group of the Hungarian Academy of Sciences Szeged (LK), Szeged, Hungary

ABSTRACT Lipedema was first described by Allen and Hines more than seven decades ago but Lipedema is a disproportional obesity its pathomechanism has remained enigmatic due to unknown pathomechanism. Its major (1). It is a disproportional obesity that nearly hallmark is frequent hematoma formation always affects women and is characterized by related to increased capillary fragility and bilateral, symmetrical, biker’s hosiery-shaped reduced venoarterial reflex. Beyond micro- fatty swelling of the legs with sparing of the angiopathy, both venous and lymphatic feet (2-4). Arms are also commonly affected dysfunction have also been documented. by adipose hypertrophy. However, during However, arterial circulation in lipedema has progression, the characteristic “stove-pipe” not been examined, and therefore we explored legs emerge and as a result of orthostatic aortic elastic properties by echocardiography. prolongation, a sharp demarcation between Fourteen women with and 14 without normal and abnormal tissue at the ankle lipedema were included in the study. Each occurs with the filling of retromalleolar subject consented to blood pressure measure- sulcus, giving the “pantaloon” appearance ment, physical examination, and transthoracic (Fig. 1). echocardiography. Aortic stiffness index (ß), The skin is usually normal in texture and distensibility, and strain were evaluated from appearance, without the dermal thickening or aortic diameter and blood pressure data. induration common in . At most, Mean systolic (30.0 ± 3.2 vs. 25.5 ± 3.6, one in ten women or postpubertal girls may P<0.05) and diastolic (27.8 ± 3.3 vs. 22.3 ± 3.1) exhibit involvement to some degree (5). aortic diameters (in mm) and aortic stiffness Lipedema is presumably associated with index (9.05 ± 7.45 vs. 3.76 ± 1.22, P<0.05) genetic background, but, to date, none of the were significantly higher, while aortic strain examined genes have proved to be responsible (0.082 ± 0.04 vs. 0.143 ± 0.038, P<0.05) and for the phenotypic appearance (6). It is easily distensibility (2.24 ± 1.07 vs. 4.38 ± 1.61, distinguishable from lymphedema and P<0.05) were significantly lower in lipedema- phleboedema by clinical features (2,4,5), and tous patients compared to controls. Thus, hormonal abnormalities can commonly be lipedema is characterized with increased detected (2,4,5). Lipedema is often combined aortic stiffness. with morbid obesity, although lipedematous tissue hardly or never responds to diet and Keywords: lipedema, aortic stiffnes, forced weight loss. Lipedema patients nearly echocardiography, blood pressure always complain of spontaneous or mild

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY. 72

visualized by lymphscintigraphy, and this worsens with aging (10). As soon as overload of lymphatics occurs, lymph vessels are unable to maintain their function and altered microcirculation leads to impaired lymph transport capacity and accumulation of lymph fluid. Lymph stasis induces fibrosis leading to non-pitting characterized by Stemmer’s sign (5). Despite alterations in capillaries, lymphatics, and veins, the function and morphology of large arteries have not previously been studied and therefore, not been examined in relation to lipedema. The markedly reduced elasticity of the skin and the poor venoarterial reflex in conjunction with robust fatty enlargement also substan- tially weakens venous calf pumping activity leading to further impaired venous and lymphatic function. Massively enlarged adipose tissue and accompanying edema may increase vascular resistance affecting arterial function. Some data suggest that body weight and fat distribution are related to higher arterial stiffness (12), and these data have stimulated our examination of aortic Fig. 1. Lipedema: clinical appearance. elasticity in lipedema. Two-dimensional transthoracic echocardiography (2DE) is a valuable and injury evoked pain upon palpation that non-invasive method for assessment of aortic worsens with aging. Another hallmark is stiffness index (ß), distensibility, and strain as frequent hematoma formation due to even a reliable characteristics of arterial elasticity minor trauma (2,4,5). This feature is depicted (13,14). The current study was designed in increased capillary fragility (CF) and to measure these parameters of elasticity in partially with decreased venoarterial reflex women with and without lipedema. (VAR) (7,8). Lipedema is accompanied by local lymphatic and venous abnormalities PATIENTS AND METHODS (e.g., irregularly dilated lymphatics, lymphatic microaneurysms, and varicosity), Study Population however major venous dysfunction is rarely found (9-11). The study included 14 women with and The peculiar enlargement of subcutaneous 14 without lipedema. All of them were fat is linked with microangiopathy and referred to our outpatient clinic for routine altered microcirculation leading to increased cardiological examination, whereas lipedema permeability and protein-rich fluid extra- patients were first seen at the lymphedema vasation, both of which further enhance the outpatient care unit in the Department of amount of lymph produced. Therefore, in Dermatology and Allergology (Table 1 early stages, an increased lymph flow may be displays patient demographics). All patients

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY. 73

TABLE 1 Patient Demographics

underwent physical examination, 2DE, and three consecutive measurements was accepted blood pressure measurement. Patients with as the blood pressure measurement. coronary or valvular heart disease, atrial fibrillation or other arrhythmias, heart Transthoracic Echocardiography failure, unstable angina pectoris, or acute myocardial infarction were excluded. Body All subjects underwent a complete mass index (BMI) was calculated by dividing 2DE and Doppler study using a Toshiba the participant’s weight in kilograms by the Powervision 8000 and Aplio echocardio- square of his/her height in meters. The graphy equipments (Toshiba, Tokyo, Japan) following classification was used: overweight in the left lateral decubitus position from (BMI 25-29.9 kg/m2), grade 1 obesity (BMI multiple windows. 2-dimensional echo 30-34.9 kg/m2), grade 2 obesity (BMI 35-39.9 cardiography-guided M-mode tracings were kg/m2) and grade 3 obesity (BMI greater used to measure systolic and diastolic than 40 kg/m2) (12). Informed consent was ascending aortic diameters (SD and DD, obtained from each patient, and the study respectively) using the leading-to-leading protocol conformed to the ethical guidelines edge technique at a level of 3 cm above the of the 1975 Declaration of Helsinki, as aortic valve from parasternal long-axis view, reflected in a prior approval by the according to a method described previously institution’s Human Research Committee. (14). Gain, depth, and sector angles were set in an individualized manner to provide the Blood Pressure Measurement best measurement. The SD and DD were measured at the maximum anterior motion of After clinical and demographic data the aorta and at the peak of QRS complex on recording, systolic and diastolic blood the simultaneously recorded ECG, pressures (SBP and DBP, respectively) were respectively (Fig. 2). measured in the supine position with a mercury cuff sphygmomanometer from the Evaluation of Aortic Stiffness, Distensibility left arm after 10 min of rest. The first and the and Strain fifth Korotkoff sounds were taken for the SBP and DBP. Stimulants were not used or Aortic stiffness index (ß) was used as consumed during the 30 minutes prior to the a characteristic of aortic elasticity, which blood pressure measurements. The average of represents the slope of the exponential

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY. 74

of the vascular wall. The following formula was used to calculate: (ß) = ln(SBP/DBP)/(DD/DD), where DD is the pulsatile change in aortic diameter (SD minus DD) and ‘ln’ is the natural logarithm (15). The inverse of stiffness is compliance (distensibility), which describes the ease of systolic aortic expansion and is calculated as follows: 2 x (SD - DD)/(SBP-DBP) x DD). Aortic strain expressed as a percentage change of the aortic root was calculated as (SD-DD)/DD.

Statistical Analysis

Data are reported as means ± standard deviation; 95% confidence limits are also included. We applied Mann-Whitney and Fisher’s exact tests using SPSS 12.0 software. A probability value P<0.05 was accepted as statistically significant.

RESULTS

Patient Characteristics

There was no significant difference between patient and control groups as to age (40 ± 9 vs. 36 ± 7 years, P>0.05) and mean BMI values (27.6 ± 1.7 vs. 27.2 ± 3.3 kg/m2, P>0.05). None had relevant risk factors including lipid metabolism disturbance, diabetes mellitus, or anemia (Table 1).

Transthoracic Echocardiography

Mean systolic and diastolic aortic diameters and aortic stiffness index were significantly higher, while aortic strain and distensibility were considerably lower in Fig. 2. Measurement of systolic (DS) and diastolic lipedematous patients compared to control (DD) diameters of the ascending aorta are shown on the M-mode tracing obtained at a level 3 cm above subjects (Table 2). Individual systolic blood the aortic valve. pressures are displayed in Fig. 3 and aortic stiffness indices are presented in Fig. 4. function relating the relative arterial pressure DISCUSSION and the distention ratio of the artery and characterizes the entire deformation behavior The present study has clearly shown that

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY. 75

TABLE 2 Echocardiographic Data, Aortic Dimensions and Elastic Properties in Lipoedema Patients and Controls

Lipedema Group Controls

LV end-diastolic diameter (mm) 49.4 ± 3.4 48.4 ± 3.0 LV end-systolic diameter (mm) 30.8 ± 2.8 30.1 ± 3.5 Interventricular septum (mm) 9.4 ± 1.1 9.4 ± 0.8 LV posterior wall (mm) 9.0 ± 1.0 9.1 ± 1.0 LV ejection fraction (%) 67.4 ± 5.9 68.1 ± 5.4 Aortic systolic diameter (SD, mm) 30.0 ± 3.2 * 25.5 ± 3.6 Aortic diastolic diameter (DD, mm) 27.8 ± 3.3 * 22.3 ± 3.1 SD-DD (mm) 2.20 ± 0.92 * 3.21 ± 0.87 Systolic blood pressure (SBP, mm Hg) 135.4 ± 9.9 129.6 ± 13.9 Diastolic blood pressure (DBP, mm Hg) 79.1 ± 12.4 77.9 ± 8.6 SBP-DBP (mm Hg) 56.3 ± 12.8 51.7 ± 13.1 Aortic strain 0.082 ± 0.040 * 0.143 ± 0.038 Aortic distensibility (cm2/dynes 10-6) 2.24 ± 1.07 * 4.38 ± 1.61 Aortic stiffness index (ß) 9.5 ± 7.45 * 3.76 ± 1.22

* p <0.05 vs. controls; Abbreviations: LV: ejection fraction aortic stiffness is increased in lipedematous functional (decreased aortic elasticity) vascu- patients compared to healthy individuals lar abnormalities, where aortic enlargement within a similar age group suggesting early may represent an adaptation process to vascular remodeling. Increased arterial accommodate the increased blood volume. stiffness is an important risk factor for Wildman et al found that excess weight cardiovascular mortality in distinct diseases begins to affect the vascular system at a very (16). The evidence demonstrated that early stage of vascular aging (17). We have pulsatile changes in ascending aortic vessel recently shown that obesity is associated with diameter can be indirectly registered during aortic enlargement and increased stiffness routine 2DE (13) instead of using direct inva- regardless of the age of patients (12). sive methods. Aortic elastic properties rely on Various mechanisms can be attributed aortic and blood pressure data. These non- to obesity-linked aortic stiffening including invasive measurements have been confirmed visceral adipose tissue throughout insulin as the determinants of aortic distensibility, resistance development (18). Hyperinsulin- with a high degree of accuracy (13). emia can be associated with increased Lipedema is a peculiar form of adipose sodium reabsorption (19), a stimulated tissue deposition in a disproportional fat sympathetic nervous system (20), and accumulation in a biker’s hosiery shape and atherosclerosis (21,22). In the obesity-related is complicated by venous and lymphatic insulin resistant state, the endothelium insufficiency and also microangiopathy. dependent vasodilator effects of insulin is Generalized obesity is known to be associated weakened (23,24). Higher levels of with structural (altered aortic size) and nonesterified fatty acids have been found to

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY. 76

Fig. 3. Individual systolic blood pressures of lipedema patients and controls. be associated with central obesity by Several theories have been postulated increasing α-adrenergic reactivity, vascular regarding the etiology of lipedema. Földi and tone, and blood pressure (25). Increased BMI Földi have proposed that microangiopathy in has been associated with permanent low- the area of the affected adipose tissue sets off grade inflammation and concomitantly the condition leading to increased perme- increased expression of proinflammatory ability to proteins (5,8). Hypoxia is known to cytokines resulting in wall stiffening (26,27). be a major induction factor for angiogenesis One of the major hallmarks of lipedema and, in the eye, pathological angiogenesis in is excessive bruising. Hematoma formation is the retina leads to catastrophic loss of vision presumably triggered by increased CF and in retinopathy of maturity, diabetic retino- possibly by impaired venoarterial reflex pathy. The principal feature of these newly (VAR) (7,8). Capillary fragility measurement formed capillary vessels in these disorders is with a vacuum suction chamber, the Parrot’s also fragility. Angiogenesis is controlled by angiosterometer, has clarified remarkable several factors, including vascular endothelial CF in the background of frequent hematoma growth factor (VEGF), and accordingly, formation (7). Decongestive lymphatic abnormally high serum VEGF levels might (DLT) in combination with inter- be predicted in the presence of intensive mittent pneumatic compression (IPC) angiogenesis and concomitant CF. A study significantly decreased CF. examining effects of shock-wave therapy in

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY. 77

Fig. 4. Individual aortic stiffness indices of lipedema patients and controls. patients with lipedema or cellulite found explanation for the increased aortic stiffness. nearly 4-fold higher plasma mean VEGF Lipedematous adipose tissue as in generalized levels at baseline compared to non- morbid obesity may enhance peripheral lipedematous individuals (28). Other vascular resistance causing increasing aortic observations relate to immunohistochemical stiffness (12). Additionally, at the microvas- findings (29). In lipedema, adipocytes cular level, localized hypertension accom- undergo necrosis and are scavenged by panying arteriolar remodeling and capillary macrophages. Hypoxia induced by excessive hyperpermeability can lead to microangio- adipose hypertrophy has been proposed to pathy interacting with macrovascular cause adipose metabolic dysfunction and function. Expansion of the lipedematous the production of adipose tissue cytokines microvascular network and its fragility might related to inflammatory reaction might also be explained as a compensatory response to be operative in lipedema. Furthermore, constant stimuli from local tissue pressure adipocyte enlargement and Ki67 and CD34 and vascular resistance. Thus, the known positivity were also observed, which are interaction of macro- and microcirculations associated with cell proliferation, and adipose observed in other disorders could account stem/progenitor cells, respectively. for the coexistence of raised CF and aortic Observed and suspected pathological stiffness in lipedema (30). features of lipedema suggest a possible To our knowledge, this is the first

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY. 78 measurement of aortic stiffness in lipedema Funktionslymphszintigraphie beim Lipödem patients, and we found remarkable und Lipolymphödem. LymphForsch 4 (2000): alterations in aortic elastic properties. 74-77. 11. Amann-Vesti, BR, UK Franzeck, A Bollinger: Microlymphatic aneurysms in patients with ACKNOWLEDGMENTS lipedema. Lymphology 34 (2001), 170-175. 12. Nemes, A, H Gavallér, E Csajbók É, et al: The authors gratefully acknowledge the Obesity is associated with aortic enlargement and increased stiffness: An echocardiographic skilled assistance of the nursing staff of the study. Int. J. Cardiovasc. Imaging 24 (2008), nd Echocardiographic Laboratory of the 2 165-171. Department of Medicine at the Cardiology 13. Stefanadis, C, C Stratos, H Boudoulas, et al: Center, University of Szeged, Szeged, Distensibility of the ascending aorta: Hungary. The authors also wish to thank Comparison of invasive and non-invasive techniques in healthy men and in men with Professor Dr. Hugo Partsch for his invaluable coronary artery disease. Eur. Heart J. 11 help in manuscript text and figure (1990), 990-996. constructions. 14. Nemes, A, TW Galema, ML Geleijnse, et al: Attila Nemes is supported by the EGT Aortic valve replacement for aortic stenosis is Norway Grant and Financial Mechanism - associated with improved aortic distensibility at long-term follow-up. Am. Heart J. 153 OTKA (NNF-78795). (2007), 147-151. 15. Henry, WL, A DeMaria, R Gramiak R, et al: REFERENCES Report of the American Society of Echo- cardiography Committee on nomenclature and standards in two-dimensional echocardio- 1. Allen, EV, EA Hines: Lipedema of the legs: graphy. Circulation 62 (1980), 212-217. A syndrome characterized by fat legs and 16. Nemes, A, T Forster, M Csanády, et al: orthostatic edema. Mayo Clin. Proc. 15 Indices of aortic distensibility and coronary (1940), 184-187. flow velocity reserve in patients with different 2. Langendoen, SI, L Habbema L, TE Nijsten, grades of aortic atherosclerosis. Int. J. et al: Lipoedema: from clinical presentation Cardiovasc. Imaging 20 (2004), 271-277. to therapy. A review of the literature. Br. J. 17. Wildman, RP, RH Mackey, A Bostom, et al: Dermatol. 161 (2009), 980-986. Measures of obesity are associated with 3. Szolnoky, G, L Kemény: Lipoedema: From vascular stiffness in young and older adults. clinical presentation to therapy. Further Hypertension 42 (2003), 468-473. aspects. Br. J. Dermatol. 162 (2010), 889. 18. Brunzell, JD, JE Hokanson: of 4 Fife, CE, EA Maus, MJ Carter: Lipedema: a central obesity and insulin resistance. frequently misdiagnosed and misunderstood Diabetes Care 22(Suppl 3) (1999), C10-C13. fatty deposition syndrome. Adv. Skin. Wound 19. Stenvinkel, P, J Bolinder, A Alvestrand: Care 23 (2010): 81-92. Effects of insulin on renal haemodynamics 5. Földi, M, S Kubik: Textbook of Lymphology. and the proximal and distal tubular sodium Urban and Fischer Publisher, 2005. handling in healthy subjects. Diabetologia 35 6. Child, AH, KD Gordon, P Sharpe, et al: (1992), 1042-1048. Lipedema: An inherited condition. Am. J. 20. Young, JB: Effect of experimental hyper- Med. Genet. A. 152 (2010), 970-976. insulinemia on sympathetic nervous system 7. Szolnoky, G, N Nagy, RK Kovács, et al: activity in the rat. Life Sci. 43 (1988), 193-200. Complex decongestive physiotherapy reduces 21. Ulrich, P, A Cerami: Protein glycation, capillary fragility in lipedema. Lymphology diabetes, and aging. Recent Prog. Horm. Res. 41 (2008), 161-166. 56 (2001), 1-21. 8. Földi, E, M Földi: A comparative study to 22. DeFronzo, RA, E Ferrannini: Insulin resis- measure veno-arterial reflex in lipedema. In: tance. A multifaceted syndrome responsible Lipoedema and Cellulitis. Viavital Publisher for NIDDM, obesity, hypertension, dyslipi- 2001: 80-86. demia, and atherosclerotic cardiovascular 9. Harwood, CA, RH Bull, J Evans, et al: disease. Diabetes Care 14 (1994), 173-194. Lymphatic and venous function in lipoedema. 23. Kahn, BB, JS Flier: Obesity and insulin Br. J. Dermatol. 134 (1996), 1-6. resistance. J. Clin. Invest. 106 (2000), 473-481. 10. Brauer, WJ: Altersbezogene 24. Arcaro, G, A Cretti, S Balzano, et al: Insulin

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY. 79

causes endothelial dysfunction in humans: 29. Suga, H, J Araki, N Aoi, et al: Adipose tissue sites and mechanisms. Circulation 105 (2002), remodeling in lipedema: adipocyte death and 576-582. concurrent regeneration. J. Cutan. Pathol. 36 25. Egan, BM, G Lu, EL Greene: Vascular effects (2009), 1293-1298. of non-esterified fatty acids: Implications for 30. Feihl, F, L Liaudet, B Waeber: The the cardiovascular risk factor cluster. macrocirculation and microcirculation of Prostaglandins Leukot. Essent. Fatty Acids hypertension. Curr. Hypertens. Rep. 11 60 (1999), 411-420. (2009), 182-189. 26. Weyer, C, JS Yudkin, CD Stehouwer, et al: Humoral markers of inflammation and endothelial dysfunction in relation to Gyözö Szolnoky, MD, PhD adiposity and in vivo insulin action in Pima Indians. Atherosclerosis 161 (2002), 233-242. Department of Dermatology and 27. Visser, M, LM Bouter, GM McQuillan, et al: Allergology Elevated C-reactive protein levels in University of Szeged overweight and obese adults. JAMA 282 P.O. BOX 427 (1999), 2131-2135. H-6720 Szeged, Hungary 28. Siems W, T Grune, P Voss, et al: Anti- fibrosclerotic effects of shock wave therapy in Tel: +36-20-326-6161 lipoedema and cellulite. Biofactors 24 (2005), Fax: +36-62-545-954 275-282. E-mail: [email protected]

Permission granted for single print for individual use. Reproduction not permitted without permission of Journal LYMPHOLOGY.