Galli, Massimo; Veglia, Fabrizio; Angarano, Gioacchino; Santambrogio, Sara; Meneghini, Elena; Gritti, Francesco; Cargnel, Antonietta; Mazzotta, Francesco; Lazzarin, Adriano
From the Institute of Infectious and Tropical Diseases, University of Milan (Drs Galli and Santambrogio); Institute for Scientific Interchange (ISI) Foundation, Turin (Dr Veglia); Institute of Infectious Diseases, University of Foggia (Dr Angarano); IRCCS San Raffaele, Milan (Drs Meneghini and Lazzarin); Institute of Infectious Diseases, University of Bologna (Dr Gritti); I Division of Infectious Diseases, L. Sacco Hospital, Milan (Dr Cargnel); and Institute of Infectious Diseases, S.M. Annunziata Hospital, Firenze, Italy (Dr Mazzotta).
Received for publication February 18, 2003; accepted June 27, 2003.
Supported by the Istituto Superiore di Sanità, Rome, Italy (grant ISS 30C.53 and grant ISS 39C.4) and by the Italian Ministry of Health (grant ICS 030.5/RF99.55).
Address correspondence and reprint requests to Massimo Galli, Institute of Infectious Diseases and Tropical Medicine, Università di Milano–Ospedale Luigi Sacco, Via G.B. Grassi 74, 20157 Milano, Italy. E-mail: email@example.com
A dipose tissue abnormalities (ATAs) involving many body regions have been observed in a number of HIV-1–infected people receiving antiretroviral therapy (ART). 1–3 It has been reported that gender influences the ATA pattern, and a magnetic resonance imaging study has shown that truncal adiposity rather than subcutaneous fat wasting is the dominant manifestation in women on ART. 4 These observations, which are also supported by other studies of body composition, 5–7 have generated the opinion that the prevalent pattern of ATA in women on ART can be defined as an “android” body habitus. 5
The HIV epidemic in industrialized countries has involved more men than women, and women are therefore generally underrepresented in cohort studies of ART. The aim of this cross-sectional investigation was to obtain more information concerning gender-related differences in ATA patterns by comparing the prevalence of ATAs in any body site and their risk correlates in a large series of HIV-infected patients, almost 30% of whom were women.
The Lipodystrophy Italian Multicentre Study (LIMS) was designed to include the outpatients who consecutively attended routine visits at 6 infectious disease departments in 4 Italian cities (Milan, Bologna, Florence, and Bari) during a fixed 30-day period. The exclusion criteria were an age younger than 18 years, refusal or inability to provide informed consent, a previous diagnosis of AIDS wasting syndrome or AIDS dementia complex, the administration of steroids or antineoplastic chemotherapy in the 12 months preceding enrollment, or the initiation of a hypocaloric diet during the same period.
The demographic and clinical data on each patient, including all of the antiretroviral drugs and their combinations taken after starting ART, and the time of and reason for any discontinuations, were entered in a centralized database. The patients were asked to report any changes in their bodily appearance occurring since the start of ART (or during the previous 2 years in the case of untreated patients), and the physicians were asked to describe such changes accurately. The alterations reported by the patients but not confirmed by the physicians were not considered in the analysis.
The considered ATAs included fat accumulation in the cervical (bull neck), supraclavicular, and interscapular regions (buffalo hump), as well as the breasts, trunk, and abdomen; fat loss in the face, arms, lower limbs, and glutei; and lipomas (defined as circumscribed, well-delimited, round-shaped adipose tissue masses). The Marrakesh classification 8 was used to define the ATAs as type 1 (pure lipoatrophy), type 2 (pure lipohypertrophy), or type 3 alterations (the simultaneous presence of fat loss and fat accumulation in different body regions).
The odds ratios of presenting with ATAs in each body region, and the different types in men and women, were assessed by means of logistic regression. The risk ratios were adjusted for age (per 1-year increase), the Centers for Disease Control stage of HIV disease, plasma HIV RNA load (detectable vs. undetectable) and CD4 cell count (per 50-cell/μL increase) at enrollment, overall ART duration (per 1-month period), the largest number of drugs used in an ART combination (none, 1 nucleoside reverse transcriptase inhibitor [NRTI], 2 NRTIs, 2 NRTIs plus 1 nonnucleoside reverse transcriptase inhibitor [NNRTI], 2 NRTIs plus 1 protease inhibitor (PI), 2 NRTIs plus 2 PIs, and other combinations (such as ≥3 drugs including a PI or ≥2 drugs including an NNRTI), and the use of stavudine (d4T).
The study included a total of 2258 patients, 673 of whom were women (29.8%). A higher percentage of men had reached a more severe stage of HIV disease (P < 0.01 in the test for trend). The women had a median age of 35.0 years (interquartile range [IQR] 32–39), and the men a median age of 37.0 years (IQR 34–42). Forty-seven percent of the men (n = 746) and 36% of the women (n = 243) were current or past drug users.
At the time of the manifestation of an ATA (or at the enrollment visit in the case of patients without any signs of an ATA), treatment distribution was significantly different between men and women (χ2P < 0.001). In particular, the percentages of never-treated patients, ART interruptions, and suboptimal treatments with 2 NRTIs were higher among the women, whereas the men had more frequently received treatment including PIs. There was no significant between-gender difference in the percentage of patients who were receiving d4T at enrollment or had received it before enrollment (Table 1).
The median duration of ART was 2.8 years among the men (IQR 1.32–5.45) and 2.9 years among the women (IQR 1.35–5.69), with no significant between-gender difference even after adjusting for the ART combinations.
Confirmed ATAs were observed in 750 patients (33.2%), 282 of whom were women. Confirmed breast enlargements in men were included in the analysis regardless of the possible presence of glandular hypertrophy; all but 1 of these men also had at least 1 ATA in another body region.
ATAs in any body region (without exception) were more frequent in the women than in the men, who had a significantly lower adjusted risk for them in all cases. The risk of presenting with circumscribed lipomas was not significantly different between genders. The risk of pure (type 2) and combined hypertrophy (type 3) was also significantly lower in men, whereas there was no significant between-gender difference in the risk of developing pure atrophy (type 1) (Table 2).
The results of this large cross-sectional study clearly show that women are at higher risk of developing ATAs than men. This increased risk emerged despite their significantly less intensive ART treatment (a smaller percentage of women had received PIs), lower stage HIV disease, and the fact that male breast enlargements (probably including many cases of glandular hypertrophy without any significant fat accumulation) were arbitrarily attributed to fat accumulation and considered in the analysis. Furthermore, there was no significant between-gender difference in the percentage of patients receiving d4T (a drug reportedly associated with an increased risk of lipoatrophy) 2,3,9 nor in the overall duration of ART adjusted for the combination used.
It is interesting to note that there was a marked difference in the ATA profiles of the men and women. Type 1 ATAs (pure lipoatrophy) were the most frequent alteration in men and the least frequent in women, who were prevalently affected by type 3 ATAs (combined forms). It is also worth noting that the risk of developing pure lipoatrophy was almost equal between the 2 genders.
Previous longitudinal studies of PI-treated patients, 2,3 and PI-naive patients receiving 2 NRTIs, 10 have shown the fat accumulation (and combined forms, when considered separately) are always significantly more frequent in women, whereas their risk of pure fat atrophy is less than, or not significantly different from, that of men. However, 2 French studies did not find a significantly increased risk of ATAs in women but reported a significantly increased risk of lipoatrophy in men. 11,12 It cannot be excluded that the differences between the French and Italian/Spanish studies may have been at least partially due to differences in diet, personal behaviors, or case selection, but the fact that our study included more women than the French studies provided us with a more favorable opportunity of identifying between-gender differences in ATA distribution.
Studies of body composition 5–7 have suggested that women with lipodystrophy tend to develop an android body habitus characterized by increased trunk adiposity. Although not supported by instrumental examinations, our data reveal a complex pattern of alterations, with a striking degree of fat accumulation localized in the breast and accompanied or not by fat loss in the arms and glutei, as previously reported. 13
We therefore think that, at least at first presentation and in a large number of cases, the ATA picture in women cannot easily be defined as android. However, it should be borne in mind that ATAs (particularly adipomasty) in women receiving ART can significantly change over time. 14 The follow-up data from the large series of women involved in the present study will provide an opportunity of investigating whether “generalized” trunk adiposity with reduced breast fat mass is a common outcome in ART-treated women initially presenting with breast enlargement.
Hormonal studies cannot explain the android pattern reported in female patients. Although increased free testosterone and luteinizing hormone–follicle-stimulating hormone ratios have been reported in women with lipodystrophy, 15 no significant difference in plasma testosterone was found in a series of HIV-infected women on ART with and without ATAs including breast fat accumulation. 13 Furthermore, lipodystrophy in women is not generally associated with amenorrhea, hirsutism, or other signs of endocrine imbalance, suggesting a prominent role of altered male hormone incretion.
Like all studies based on clinical observations, ours may be biased by confounding factors due to subjective judgments, including the fact that women may be more likely to complain of ATAs than men. However, we paid particular attention to any change in body appearance in a setting in which all of the patients were well informed, and both the men and women were very attentive toward lipodystrophy.
In conclusion, at least in our Southern European setting, lipodystrophy is more frequent and more polymorphic in women than in men. This difference is not due to differences in age, the severity of HIV disease, immunologic or virologic responses to therapy, or the duration or complexity of ART. Studies of the role of hormonal mechanisms will probably provide further information concerning gender-related differences in the development of lipodystrophy.
1. Safrin S, Grunfeld C. Fat distribution and metabolic changes in patients with HIV infection. AIDS. 1999; 13:2493–2505.
2. Martinez E, Mocroft A, Garcia-Viejo MA, et al. Risk of lipodystrophy in HIV-1 infected patients treated with protease inhibitors: a prospective cohort study. Lancet. 2001; 357:592–598.
3. Galli M, Cozzi-Lepri A, Ridolfo AL, et al., for the LipoICoNa study. Incidence of adipose tissue alterations in first-line antiretroviral therapy: the LipoICONA study. Arch Intern Med. 2002; 162:2621–2628.
4. Kotler DP, Lopes J, Engelson ES, et al. Interaction among sex, HIV infection and fat redistribution. AIDS Reader. 2000; 10:589–594.
5. Hadigan C, Meigs JB, Corcoran C, et al. Metabolic abnormalities and cardiovascular disease risk factors in adults with human immunodeficiency virus infection and lipodystrophy. Clin Infect Dis. 2001; 32:130–139.
6. Hadigan C, Miller K, Corcoran C, et al. Fasting hyperinsulinemia and changes in regional body composition in human immunodeficiency virus-infected women. J Clin Endocrinol Metab. 1999; 84:1932–1937.
7. Falutz J, Muurahainen N, Workman C, et al. Gender is associated with increased body size but not fat depletion in patients with HIV-associated adipose redistribution syndrome (HARS). Paper presented at: XIII International AIDS Conference; July 9–14, 2002; Durban, South Africa. Abstract WePeB 4273.
8. Galli M, Polo R, Saint-Marc T, et al. for the ART-associated Lipodystrophy European Collaborative Study (ALECS) group. Guidelines to describe morphologic and metabolic alterations (MMA) under ART: the Marrakech classification. Nutr Metab Disorders HIV Infect. 2002; 1:25–29.
9. Saint Marc T, Partisani M, Poizot-Martin I, et al. A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS. 1999; 13:1659–1667.
10. Galli M, Ridolfo AL, Adorni F, et al. Body habitus changes and metabolic alterations in protease inhibitor-naive HIV-1-infected patients treated with two nucleoside reverse transcriptase inhibitors. J Acquir Immune Defic Syndr. 2002; 29:21–31.
11. Thiebaut R, Daucourt V, Mercie P, et al. Lipodystrophy, metabolic disorders, and human immunodeficiency virus infection: Aquitaine Cohort, France, 1999. Groupe d'Epidemiologie Clinique du Syndrome d'Immunodeficience Acquise en Aquitaine. Clin Infect Dis. 2000; 31:1482–1487.
12. Saves M, Raffi F, Capeau J, et al. The Antiproteases Cohort (APROCO) Study Group. Factors related to lipodystrophy and metabolic alterations in patients with human immunodeficiency virus infection receiving highly active antiretroviral therapy. Clin Infect Dis. 2002; 34:1396–1405.
13. Gervasoni C, Ridolfo AL, Trifiro G, et al. Redistribution of body fat in HIV-infected women undergoing combined antiretroviral therapy. AIDS. 1999; 13:465–471.
14. Galli M, Ridolfo AL, Adorni F, et al. Correlates of risk of adipose tissue alterations and their modifications over time in HIV-1 infected women treated with antiretroviral therapy. Antivir Ther. 2003. In press.
15. Hadigan C, Corcoran C, Piecuch S, et al. Hyperandrogenemia in human immunodeficiency virus-infected women with the lipodystrophy syndrome. J Clin Endocrinol Metab. 2000; 85:3544–3550.