Immediate versus Delayed Surgical Intervention for Reconstructive Therapy of HIV-Associated Facial Lipoatrophy: A Randomized Open-Label Study

Abstract

We assessed the safety and efficacy of reconstructive therapy with facial fillers for the treatment of HIV-associated facial lipoatrophy (FLA) through a randomized, controlled, open-label single-center study. A total of 134 HIV-infected patients with severe FLA were randomly assigned to receive immediate (67 patients) or delayed (67 patients) facial injections of poly-l-lactic acid (PLA) or polyacrylamide gel (PAIG). Outcome measures included changes in physician and patient FLA severity scale, adverse events, and changes in health-related quality of life (HRQoL) and anxiety using validated measures. The mean average study follow-up was 27 weeks for the immediate and 25 weeks for the delayed subjects. Adverse events were mild and resolved after a mean of 4 days. Compared to patients randomized to the delayed treatment group, patients assigned to the immediate treatment group had significantly lower physician-rated (0.0 versus −3.0; p < 0.0001) and patient-rated (0.1 versus −1.8; p < 0.0001) FLA severity scores. By contrast, measures exploring HRQoL and anxiety did not show any significant difference between patients randomized to the immediate and deferred groups. Reconstructive therapy with facial fillers was effective and safe and led to significant improvements in FLA severity. However, no significant gains in HRQoL, relational and psychological consequences of body changes, and anxiety-related concerns were observed. Studies should be performed to identify patients who could maximally benefit from filling interventions for FLA.

Introduction

T he introduction of highly active antiretroviral therapy (HAART) for the treatment of HIV-infected persons has dramatically reduced mortality and morbidity,¹ making health-related quality of life (HRQoL) improvements an important goal of therapy.^2,3

Unfortunately, a significant proportion of patients receiving HAART may develop morphological abnormalities generally termed lipodystrophy syndrome, consisting of either lipoatrophy or visceral fat accumulation, or both.^4

–7 Peripheral fat wasting and lipoatrophy are particularly worrisome since they may stigmatize patients negatively, affecting their lifestyle.⁸

Incomplete knowledge of syndrome etiology, pathogenesis, and evolution has made it difficult to establish suitable therapeutic strategies to confront all aspects related to abnormal body fat distribution.^9
–11

Facial lipoatrophy (FLA) is one of the most distressing and stigmatizing expressions of the antiretroviral-associated lipodystrophy syndrome. It may deeply affect the quality of life of patients, rendering them more likely to be identified as HIV positive, thus leading to decreased self-esteem and social withdrawal.^12
–14

Despite the improved knowledge on FLA etiology and risk factors, successful medical treatments for the management of this side effect of HIV are still troublesome.¹⁵

Possible interventions include changes in living habits (diet and physical exercise) and HAART modification, such as switching from stavudine or zidovudine to abacavir or tenofovir. However, FLA reversal after drug switching is gradual, incomplete, and may expose the patient to risks of new toxicity.¹⁶ Furthermore, other pharmacological treatments for lipoatrophy, such as rosiglitazone, pioglitazone, and hormones, have not been proven effective and may induce new complications.^16
–18

Recently, surgical interventions for FLA have been increasingly considered, as they appear to offer patients immediate psychological relief.¹⁶ Among the approved facial fillers, favorable results have been reported with poly-l-lactic acid (PLA), a biocompatible, biodegradable, synthetic polymer from the alpha-hydroxy-acid family, and with polyacrylamide gel (PAIG), an immunologically inactive filler that is permanent, nonresorbable, nonallergenic, and physically and chemically stable. However, limited randomized studies on surgical interventions for FLA have been published to date. Moreover, the effects of such interventions on patient's HRQoL are poorly understood.

For these reasons, we performed a randomized open-label study of immediate versus delayed surgical intervention for the correction of HIV-associated FLA. This study was mainly aimed at evaluating the safety and efficacy of facial filler injections in the treatment of HIV-related FLA as assessed both by the treating surgeon and by the patient. Moreover, we assessed the effects of facial filler injections on HRQoL, on relational and psychological consequences of body changes, and on anxiety-related concerns.

Materials and Methods

Study design

The study was a randomized, controlled, and open-label study of immediate versus delayed treatment, with facial filler injections for HIV-positive people with FLA. Participants were randomized in a 1:1 ratio (67 immediate treatment group and 67 delayed treatment group). People randomized in the immediate treatment group received the first injection with facial fillers at baseline. Thereafter, one to four new injections with facial fillers were repeated. Subjects were scheduled to be monitored for a 24-week period. Participants in the delayed group were scheduled to receive the first injection with facial fillers at week 24 from baseline. The intention of this design was to analyze the study data for a 24-week period from baseline in both treatment groups.

Participants

Between March 2005 and July 2007 HIV-infected patients were enrolled with the following criteria: 18 years of age or older; HIV-related lipodystrophy syndrome with severe FLA, eligible on the basis of physician's recommendation for corrective surgery; CD4 count > 100/mm³; HIV-RNA < 1000 copies/ml; and stable HAART therapy for at least 6 months. Exclusion criteria included pregnancy, previous filling surgery, current anticoagulation or cortisone treatments, AIDS dementia complex, nonstable diabetes mellitus, skin diseases (such as herpes, psoriasis, and any dermatitis in an active phase), dental treatments, unsatisfactory clinical conditions, and the inability to complete the questionnaire.

All the participants in the study were enrolled by a single Italian clinic (National Institute for Infectious Diseases “L. Spallanzani,” Rome). The study was approved by the local Ethics Committee. All participants provided written informed consent before commencing the study.

Outcome measures

The primary objective of the study was to compare the change from baseline to the end of filling intervention for the immediate group or before the filling intervention for the delayed group in terms of the severity grade of the FLA assessed by physicians. The change in FLA grading score by the physicians was determined by the two plastic surgeons who undertook all the interventions, using a validated FLA severity scale that ranged from grade 1 (mild lipoatrophy) to grade 5 (most severe lipoatrophy).¹⁹ Changes in FLA severity grading were determined by the two plastic surgeons with the help of photographs obtained at baseline and at the end of the study. In case of disagreement, the less positive change in FLA severity was chosen.

Secondary objectives were to compare the two groups in terms of patient reported outcomes (PROs) including changes in FLA severity grade as determined by the patient, HRQoL measures, relationships between the lipodystrophy syndrome and the quality of life, and patient's anxiety perception.

Injection technique

Patients were mainly treated with PAIG (33 patients) (Aquamid, Contura, Soeberg, Denmark) and PLA (33 patients) (Sculptra, Aventis Pharmaceutical, Collegeville, PA) dermal fillers. Aquamid is an injectable hydrogel implant composed of 97.5% apyrogenic water bound to 2.5% cross-linked PAIG. Aquamid is a permanent, nonresorbable, immunologically inactive, nonallergenic, physically and chemically stable filler.^20,21 Sculptra is an injectable implant that contains microparticles of PLA, a biocompatible, biodegradable, synthetic polymer from the alpha-hydroxy-acid family.^22
–24 One patient received Atlean (ABR Development, Baillargues, France) because of suspected intolerance to PLA and PAIG. Atlean is a combination of tricalcium phosphate particles suspended in a hyaluronic acid gel (HA-TCP). Hyaluronic acid²⁵ occurs naturally in skin tissue, carrying water to the skin and contributing to its elasticity. It facilitates an immediate and temporary volumizing effect, whereas Beta-TCP contributes to the sustained effects of the product by stimulating endogenous synthesis of collagen.

Data collection

Demographic study data included gender, age, and route of transmission. Baseline medical information included duration of HIV disease, HIV disease stage (CDC criteria), plasma HIV RNA, CD4⁺ cell count, and the duration in years of ART therapy. FLA grade, HRQoL data, and anxiety measurements were collected at baseline and at the end of the study for both the immediate and delayed groups. The nature and incidence of adverse events were also examined at the end of filling surgery for the immediate group.

PROs measurements

The change in FLA grading score as determined by the patient was evaluated using the HIV Outpatient Study (HOPS) severity scale, which provides standardized scores: none (0), mild (1), moderate (2), and severe (3).²⁶ To measure HRQoL, the EuroQoL 5 domains (EQ-5D) and the Istituto Superiore di Sanità Quality of Life (ISSQoL) self-administered instruments were adopted.

The EQ-5D²⁷ is a widely used generic questionnaire consisting of five domains (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression) and three levels with 1 indicating “no problem,” 2 indicating “some or moderate problems,” and 3 indicating “extreme problems/impossible to do.”

The ISSQoL²⁸ is a recently validated HRQoL specific instrument appropriately designed for people living with HIV in the HAART era. This instrument includes two sections: an HRQoL Core Evaluation Form and Additional Important Areas for HRQoL. In this study only the Core Evaluation Form was used. The HRQoL Core Evaluation Form includes nine Core domains: satisfaction with quality of life (SQL), physical well-being (PW), role well-being (RW), depression and anxiety (DA), energy and vitality (EV), health distress (HD), cognitive functioning (CF), social functioning (SF), and sexual life (SL). The ISSQoL domains are scored on a 0–100 scale, with 0 indicating poorest health and 100 indicating best health.

To evaluate the relationships between the lipodystrophy syndrome and the quality of life of the study participants the ABCD (Assessment of Body Change and Distress) self-administrated questionnaire was used. The ABCD is an instrument examining body change distress as an outcome in lipodystrophy treatment trials and as a predictor of ART adherence and quality of life. For this study, the translated Italian version of ABCD was used.²⁹ The ABCD questionnaire contains 27 items and is divided into three sections. The first section is composed of six items with questions discriminating patients with/without lipodystrophy. The second section includes a question about satisfaction with one's body appearance that ranges from 1 (best satisfaction) to 5 (poorest satisfaction). The third section of the questionnaire includes 20 questions about the social and relational psychological consequences of body changes ranging from 1 (greatest adverse impact of lipodystrophy) to 5 (no impact). Responses to the 20 items are scored as a summated rating scale with 20 indicating the highest possible score and 100 indicating the lowest possible score.

To measure the person's anxiety perception the Zung self-rating anxiety scale (SAS) was used.^30

–35 The SAS is a self-administered questionnaire designed to quantify the level of anxiety for patients experiencing anxiety-related symptoms. The questionnaire is composed of 20 questions. Each question is scored on a scale of 1–4 (1, none or a little of the time; 2, some of the time; 3, a good part of the time; 4, most of the time). The total score (20 questions summated) ranges from 20 (less anxiety) to 80 (more anxiety). The standard score is obtained by dividing the total score by 80 and multiplying by 100.

Statistical analysis

Although the HRQoL instruments used in this study have undergone extensive testing for reliability and validity we retested the psychometric assumptions in our populations. Specifically, for the ISSQoL questionnaire we examined (1) reliability using Cronbach's alpha coefficient (value required: >0.70), (2) convergent validity (each item should be linearly related to its respective scale with a correlation coefficient >0.40), and (3) discriminant validity correlation of questions with their respective scales should be higher than correlations with the other scales (Δ >2 SE). The EQ-5D construct validity was tested based on a hypothesized relationship to dimensions of the ISSQoL.

We hypothesized the following ranged correlations from moderate (0.3) to strong (0.6): the EQ-5D mobility and the EQ-5D self-care with ISSQoL PW, the EQ-5D usual activity with ISSQoL PW, RW, SF, and EV, the EQ-5D pain with ISSQoL PW, RW, SF, and EV, and the EQ-5D anxiety/depression with ISSQoL DA, CF, and HD. The ABCD reliability was estimated using the Cronbach's alpha; the construct validity was tested against the ISSQoL questionnaire. We hypothesized that ABCD scores would be at least moderately correlated to all ISSQoL domains.

Baseline characteristics (demographic, clinical, HRQoL, and anxiety) were compared between immediate and delayed treatments with the Student's t test for continuous variables and with Pearsons' χ² test for the qualitative variables. The mean changes from baseline to the end of the lipofilling intervention (for the immediate group) or before the lipofilling intervention (for the delayed group) in the FLA severity scores were compared using a Student's t test.

The effect size with the confidence interval method³⁶ was used to evaluate differences between the two treatments in terms of the change from baseline to the end of the study in HRQoL measurements (ISSQoL, EQ-5D, and ABCD). The effect size d was calculated as the difference between the means divided by the pooled standard deviations (square root of the squared standard deviations average). The Cohen suggested guidelines for d are “small,” “medium,” or “large” if d is 0.20, 0.50, or 0.80, respectively.³⁷

The standardized anxiety scores were categorized in two levels: above and below the 33.8 score. The latter represents the reported normative standardized value (NSV).^30,35 The anxiety score change from baseline to week 24 was evaluated according to three different parameters: stationary (no change from baseline), improvement (a baseline value lower than the NSV becoming higher than the NSV at week 24), worsening (a baseline value higher than the NSV, becoming lower than the NSV at week 24). Statistical calculations were performed using the SAS statistical package, version 8.2 (SAS Institute, Inc., Cary, NC).

Results

Baseline characteristics

Most participants (71%) had an FLA severity grade of 3 or 4 at baseline. There were no significant differences at baseline between treatment groups with respect to demographic and clinical variables (Table 1).

Table 1.

Population Characteristics at Baseline

	Delayed treatment	Immediate treatment	All	p-value
Number, randomized	67	67	134
Gender
Female: no. (%)	10 (14.9)	10 (14.9)	20 (14.9)
Male: no. (%)	57 (85.1)	57 (85.1)	114 (85.1)	0.100^a
Age (years)
Mean ± SD (no., range)	47.6 ± 6.1 (67, 37–66)	44.5 ± 5.8 (67, 28–60)	46.1 ± 6.1 (134, 28–66)
Median	47	43	45	0.003^b
Predominant risk factor: no. (%)
Homosexual/bisexual	33 (49.3)	26 (38.8)	59 (44.0)
Intro-verious drug use	12 (17.9)	19 (28.4)	31 (23.1)
Heterosexual	22 (32.8)	22 (32.8)	44 (32.8)	0.300^a
CD4⁺/mm³ (nadir)
Mean ± SD (no., range)	175 ± 116 (67, 7–486)	189 ± 139 (67, 0–600)	182 ± 128 (134, 0–600)
Median	156	195	165	0.519^b
CD4⁺/mm³
Mean ± SD (no., range)	699 ± 390 (67, 140–2458)	664 ± 318 (67, 273–1869)	682 ± 355 (134, 140–2458)
Median	612	603	612	0.572^b
HIV-RNA cp/ml (log10)
Mean ± SD (no., range)	2.1 ± 0.8 (67, 1.7–4.4)	2.0 ± 0.7 (67, 1.7–4.5)	2.0 ± 0.7 (134, 1.7–4.5)
Median	1.7	1.7	1.7	0.397^b
Stage at randomization: no. (%)
CDC A	3 (4.5)	7 (10.4)	10 (7.5)
CDC B	36 (53.7)	36 (53.7)	72 (53.7)
CDC C	28 (41.8)	24 (35.8)	52 (38.8)	0.300^a
Time from HIV diagnosis (years)
Mean ± SD (no., range)	14.3 ± 5.3 (67, 5–22)	13.0 ± 5.0 (67, 4–23)	13.6 ± 5.1 (134, 4–23)
Median	14.0	12.0	12.0	0.139^b
Time from lipodystrophy syndrome (years)
Mean ± SD (no., range)	5.5 ± 2.6 (67, 1–15)	5.1 ± 2.2 (67, 2–11)	5.3 ± 2.4 (134, 1–11)	0.376^b
Median	5.0	5.0	5.0
Years on ART
Mean ± SD (no., range)	9.0 ± 3.7 (67, 3–20)	9.5 ± 2.9 (67, 3–19)	9.2 ± 3.3 (134, 3–20)
Median	9.0	10.0	9.0	0.472^b
PI experienced: no. (%)	55 (82.1)	60 (89.6)	115 (85.8)	0.216^a
NNRTI experienced: no. (%)	53 (79.1)	53 (79.1)	106 (79.1)	1.000^a
NRTI experienced: no. (%)	67 (100.0)	67 (100.0)	134 (100.0)
Grade of FLA by physicians: no. (%)
Grade 0	—	—	—
Grade 1	9 (13.4)	11 (16.4)	20 (14.9)
Grade 2	19 (28.4)	19 (28.4)	38 (28.4)	0.359^a
Grade 3	26 (38.8)	31 (46.3)	57 (42.5)
Grade 4	13 (19.4)	6 ( 9.0)	19 (14.2)
Grade of FLA by patient: no. (%)
Grade 0	—	—	—
Grade 1	32 (47.8)	39 (58.2)	71 (53.0)
Grade 2	35 (52.2)	28 (41.8)	63 (47.0)	0.226^a
Grade 3	—	—	—
Lipodystrophy signs frequency: no. (%)
Increase in abdominal fat	32 (47.8)	20 (29.9)	52 (38.8)	0.033^a
Increase in breast fat	14 (20.9)	13 (19.4)	27 (20.1)	0.829^a
Increase in neck fat	16 (23.9)	16 (23.9)	32 (23.9)	1.000^a
Loss of fat from legs	53 (79.1)	52 (77.6)	105 (78.4)	0.834^a

χ² test.

Student's t test.

Baseline PROs were similar for the two treatment groups for all the instruments used (Table 2).

Table 2.

Patient Reported Outcomes at Baseline

	Delayed treatment	Immediate treatment	p-value
ISSQoL Core Evaluation Form
Mean ± SD (no., range); median % floor; % ceiling
Physical well-being	66 ± 30 (67, 4–100); 75 0.0; 13.4	72 ± 25 (66, 8–100); 75 0.0; 16.7	0.248^b
Role well-being	71 ± 33 (64, 0–100); 75 7.8; 42.2	78 ± 29 (65, 0–100); 88 3.1; 47.7	0.201^b
Social functioning	76 ± 29 (64, 0–100); 88 3.1; 43.8	72 ± 30 (67, 0–100); 75 1.5; 43.3	0.376^b
Depression/anxiety	65 ± 24 (66, 11–100; 64 0.0; 9.1	65 ± 22 (67, 21–100); 61 0.0; 4.5	0.914^b
Energy/vitality	67 ± 21 (67, 6–100); 69 0.0; 11.9	70 ± 20 (67, 19–100); 69 0.0; 9.0	0.464^b
Health distress	65 ± 26 (67, 0–100); 67 3.0; 13.4	61 ± 26 (66, 0–100); 63 1.5; 9.1	0.370^b
Cognitive functioning	68 ± 25 (66, 0–100); 72 1.5; 16.7	66 ± 24 (66, 19–100); 69 0.0; 15.2	0.628^b
Sexual life	54 ± 25 (66, 5–100); 55 0.0; 1.5	54 ± 26 (67, 5–100); 60 0.0; 3.0	0.973^b
Satisfaction with quality of life	50 ± 18 (67, 0–100); 50 1.5; 1.5	55 ± 20 (66, 17–100); 50 0.0; 4.5	0.212^b
EQ-5D dimensions
Mean ± SD (no., range); median % floor; % ceiling
Mobility	1.2 ± 0.5 (64, 1–3); 1.0 1.6; 79.7	1.2 ± 0.4 (64, 1–3); 1.0 1.6; 79.7	0.413^b
No problem (%)	80	86
Some problem (%)	19	13
Extreme problem (%)	1	1	0.731^c
Self-care	1.0 ± 0.2 (65, 1–2); 1.0 0.0; 96.9	1.0 ± 0.1 (62, 1–2); 1.0 0.0; 98.4	0.591^b
No problem (%)	97	98
Some problem (%)	3	2
Extreme problem (%)	—	—	1.000^c
Usual activities	1.2 ± 0.4 (63, 1–3); 1.0 3.2; 87.3	1.2 ± 0.5 (64, 1–3); 1.0 3.1; 81.3	0.471^b
No problem (%)	87	81
Some problem (%)	10	16
Extreme problem (%)	3	3	0.675^c
Pain/discomfort	1.3 ± 0.5 (61, 1–3); 1.0 1.6; 68.9	1.5 ± 0.6 (61, 1–3); 1.0 8.2; 62.3	0.215^b
No problem (%)	69	62
Some problem (%)	29	30
Extreme problem (%)	2	8	0.278^c
Anxiety/depression	1.7 ± 0.7 (64, 1–3); 2.0 12.5; 43.8	1.7 ± 0.6 (63, 1–3); 2.0 7.9; 39.7	0.966^b
No problem (%)	44	40
Some problem (%)	44	52
Extreme problem (%)	12	8	0.531^a
Self-rating anxiety scale
Mean ± SD (no., range); median % floor; % ceiling	40.4 ± 11.2 (61, 25–80); 40.0 0.0; 4.9	42.0 ± 11.0 (63, 25–74); 41.3 0.0; 4.8	0.460^b
ABCD questionnaire
Mean ± SD (no., range); median % floor; % ceiling
Overall, how satisfied are you with how your body looks right now?	3.7 ± 1.0 (64, 1–5); 4.0 20.3; 1.6	3.4 ± 1.1 (64, 1–5); 3.5 20.3; 1.6	01.46^b
Questions 8a–t: All the questions about the social and relational psychological consequences of body changes	73.0 ± 19.0 (60, 39–100); 74.5 6.5; 0.0	75.0 ± 17.0 (62, 35–100); 74.5 5.0; 0.0	0.508^b

χ² test.

Student's t test.

Fisher exact test.

Surgery interventions and adverse events

The immediate group participants (n = 67) received an average of four filling applications (range 2–5) over a mean period of 21 weeks (range 7–55). Patients receiving PIAG, PLA, and HA + bTCP injections were 32 (47.8%), 34 (50.7%), and 1 (1.5%), respectively.

Minimal edema after 7 days (7.5%), ecchymoses after 7 days (4.5%), bleeding (4.5%), local cutaneous injury (4.5%), and subcutaneous noninflammatory nodules (1.5%) were the adverse effects observed. The mean duration of the postinjection pain was 4 days. Overall, the observed frequency of adverse events did not differ between PAIG and PLA-treated patients (data not shown).

Physician- and patient-rated FLA severity scale scores

The mean changes (from baseline to the end of filling surgery for the immediate group and just before the filling surgery for the delayed group) in FLA grading score, according to the physician evaluations, was −3.0 (SD: ± 0.9; range: from −1 to −4) for the immediate group and 0.0 (SD: ± 0.1; range: from 0 to −1) for the delayed group (p < 0.0001) (Table 3). The mean change according to the filler used was −3.2 for patients treated with PAIG and −2.7 for patients treated with PLA (data not shown in tables). No case of disagreement between the two plastic surgeons in terms of changes in FLA grading score was observed.

Table 3.

Change in Facial Lipoatrophy Severity Scale

	Delayed treatment	Immediate treatment	p-value
Physicians score		—
Mean ± SD (no., range)	0.0 ± 0.1 (67, 0–1)	−3.0 ± 0.9 (67, −4, −1)	<0.0001^a
Median	0.0	−3.0
Patient score		—
Mean ± SD (no., range)	0.1 ± 0.3 (67, 0–1)	−1.8 ± 0.8 (67, −3–0)	<0.0001^a
Median	0.0	−2.0

Student's t test.

The mean change in the patient HOPS scale from baseline to the end of filling surgery for the immediate group was −1.8 (SD: ± 0.8; range: from 0 to −3) and before the filling surgery for the delayed group was 0.1 (SD: ± 0.1; range: from 0 to −1) (p < 0.0001) (Table 3). The mean change according to the filler used was −1.7 for patients treated with PAIG and −1.9 for patients treated with PLA (data not shown in tables).

PROs measures

The psychometric properties of the PROs measures were demonstrated. Reliability for each of the ISSQoL domains was good; Cronbach's alpha was ≥0.80 for each domain. All ISSQoL domains demonstrated convergent and discriminant validity. The EQ-5D showed good construct validity. The hypothesized at least moderate correlations between the EQ-5D and ISSQoL dimensions were generally confirmed, except for self-care, for which the domains correlation was equal to 0.1. Cronbach's alpha for ABCD was 0.95. The ABCD showed the correlations ranged from 0.3 to 0.8 with all ISSQoL domains.

PROs data at the end of the study were not available for all the study participants, as some of them did not fill in the questionnaires. The PROs characteristics at baseline were similar to the ones of the 134 randomized patients. No significant differences between arms were found.

The immediate group subjects completed the questionnaires after the last filling intervention (mean 8 weeks, range 1–20 weeks), whereas the subjects of the delayed group filled in the questionnaire before the filling intervention.

The mean average study follow-up was 27 weeks (range 17–36) for the immediate and 25 weeks (range 15–35) for the delayed subjects. Baseline characteristics of the patients who filled in the questionnaires were similar to the ones reported for all the randomized patients (data not shown).

EQ-5D

On average, the EQ-5D profile of the enrolled sample of subjects showed no change during follow-up for both the immediate and delayed group. On comparing the two treatment groups at the end of the study, a “small” and not statistically significant negative effect size was observed in the “mobility” and “anxiety/depression” dimensions for the immediate group. The effect sizes for all dimensions were not statistically significant at the 95% level (Table 4).

Table 4.

Heath-Related Quality of Life Changes from Baseline

	Delayed treatment	Immediate treatment	Effect size ^a 95% IC
ISSQoL
Mean change from baseline ± SD (no., range)
median Change from baseline
Satisfaction with quality of life	4.0 ± 16.0 (50, −33–50) 0	2.0 ± 16.1 (43, −50–42) 0	−0.1 (−0.3, 0.5)
Physical well-being	6.2 ± 30.0 (49, −79–96) 0	6.2 ± 28.0 (43, −71–75) 4.2	0.0 (−0.4, 0.4)
Role well-being	10.6 ± 30.0 (45, −63–88) 0	7.1 ± 31.0 (39, −100–100) 0	−0.1 (−0.8, 0.6)
Social functioning	0.3 ± 26.0 (46, −100–50) 0	8.4 ± 26.1 (43, −38–100) 0	0.3 (−1, 1.6)
Anxiety/depression	7.1 ± 14.2 (49, −14–54) 4.0	5.0 ± 18.0 (43, −36–43) 7.1	−0.1 (−0.6, 0.4)
Energy/vitality	1.4 ± 17.2 (49, −33–63) 0	0.1 ± 16.0 (43, −44–31) 0	0.0 (−0.4, 0.4)
Health distress	10.6 ± 17.0 (50, −19–63) 6.3	10.3 ± 20.0 (43, −31–50) 6.3	0.0 (−0.4, 0.4)
Cognitive functioning	2.2 ± 13.2 (48, −25–44) 0	5.0 ± 21.4 (43, −38–56) 0	0.2 (−0.4, 0.8)
Sexual life	1.2 ± 22.0 (48, −70–50) 0	5.0 ± 20.0 (43, −35–65) 5.0	0.2 (−0.5, 0.9)
EQ-5D dimensions
Mean change from baseline ± SD (no., range)
Median change from baseline
Mobility	−0.1 ± 0.4 (45, −2–0) 0	0.0 ± 0.4 (40, −2–1) 0	0.2 (−0.5, 0.7)
Self-care	0.0 ± 0.0 (47, −0–0) 0	0.0 ± 0.2 (40, −1–0) 0	0.0 (−0.4, 0.4)
Usual activities	−0.1 ± 0.4 (45, −1–1) 0	−0.1 ± 0.5 (41, −2–1) 0	0.0 (−0.4, 0.4)
Pain/discomfort	−0.0 ± 0.6 (43, −1–2) 0	0.0 ± 0.7 (37, −2–1) 0	0.0 (−0.4, 0.4)
Anxiety/depression	−0.1 ± 0.6 (46, −2–1) 0	0.0 ± 0.5 (39, −1–1) 0	0.2 (−0.2, 0.6)
ABCD questionnaire
Mean change from baseline ± SD (no., range)
Median change from baseline
Overall, how satisfied are you with how your body looks right now?	−0.3 ± 1.1 (38, −2–4) 0.0	−0.7 ± 1.2 (38, −4–1) 0.0	0.3 (−0.1, 0.7)
All the questions about the social and relational psychological consequences of body changes	4.0 ± 10.1 (38, −15–47) 3.5	4.6 ± 14.6 (38, −32–42) 3.5	0.0 (−0.4, 0.4)

Between breaks: difference between mean change from baseline of immediate and delayed group/pooled standard deviation (square root of the average of the squared deviations).

ISSQoL

From baseline to the end of the study a mild increase for most of the ISSQoL dimensions scores (mean change from baseline) was observed in both the immediate and delayed groups. Comparing the two groups, a positive “small” effect size for cognitive function, sexual life, and social functioning of 0.2, 0.2, and 0.3, respectively, was observed for the immediate group. However, these differences were not statistically significant at the 95% level (Table 4).

ABCD

The results (in terms of mean change from baseline) reported from the ABCD questionnaire showed no noticeable change during follow-up for both treatment groups. No statistically significant differences between the immediate and delayed group at the end of the study were observed (Table 4).

SAS

In the SAS change from baseline, no statistically significant differences between the delayed and immediate group were observed (Table 5). Most participants, 73% and 86% for the delayed and immediate groups, respectively, showed no change in their status compared with baseline NSV. Only a few immediate group participants (5%) improved their anxiety level in comparison to the baseline value (lower than the NSV and becoming higher at week 24).

Table 5.

Zung Anxiety Scale Change from Baseline ^a

	Delayed treatment (N = 37)	Immediate treatment (N = 37)
Improved no. (%)	5 (13.5)	1 (2.7)
No change no. (%)	27 (73.0)	32 (86.5)
Worse no. (%)	5 (13.5)	4 (10.8)

Fisher exact test p = 0.272.

Discussion

We performed a randomized open-label study of immediate versus delayed surgical intervention with filling components for the correction of severe HIV-associated FLA. The filling components used were considered in current guidelines as a conventional strategy to correct FLA.¹⁶ Data from the literature indicate that with the use of PAIG for the reconstruction of FLA, favorable results can be obtained with good aesthetic gains.^21–22 Similarly, studies have reported that PLA injections can lead to significant improvements in HIV-related FLA.^23
–25 Thus, both PAIG and PLA are considered effective, safe, and simple treatment options for severe FLA.

In this study on HIV-infected patients with severe FLA two methods were used to define lipoatrophy severity: the physician evaluation and the patient self-assessment. The results demonstrated that both evaluations were associated with a meaningful improvement in the FLA severity grade after the filling interventions. Furthermore, filling surgery was well tolerated by most patients. Less than 8% of subjects reported mild and transient side effects. Thus, the results of our study confirm and support previous reports that filling surgery is a safe and effective procedure to correct FLA.^38

–41 The limited efficacy of alternative treatments suggests that filling interventions could represent an adjunctive strategy, together with switch in antiretroviral drugs, to correct FLA.

However, despite improvements in facial lipoatrophy grading scores, no significant differences between the immediate and delayed treatment groups were observed in terms of PROs. In particular, our study was not able to show any significant gain in HRQoL (EQ-5D and ISSQoL), in social aspects, in relational-psychological consequences of body changes (ABCD), as well as in anxiety-related concerns (SAS). The large number of PROs measures used and the large number of patients enrolled make the results of our study robust.

To date, the different HIV clinical studies addressing the consequences of surgical intervention with filling components on HRQoL, depression, and anxiety in HIV-positive persons with FLA reported conflicting results. The only randomized controlled study of immediate versus delayed treatment enrolling 31 patients published to date³⁸ showed significant improvements in HRQoL scores at week 12. However, when the week 48 analysis was carried out for all the patients as a single cohort (all subjects after the filling intervention), statistical significance was maintained only for the mental health summary score, whereas no significant differences from baseline were observed in all domains of the MOS-HIV questionnaire, suggesting only a transient and limited benefit on HRQoL.

Lafaurie et al. ³⁹ conducted a prospective open-label study on 94 patients with FLA treated with intradermal injection of PLA. In this study, the primary end-point was the self-perception of the improved lipoatrophy assessed by a visual analogue scale (VAS) and the secondary end-points were HRQoL and adverse events. Compared to baseline, although self-perception of FLA severity at the end of the treatment procedure was significantly improved, the HRQoL score remained unchanged.

A number of studies have also examined anxiety and depression in patients undergoing surgical treatment for FLA. Moyle et al. ²² conducted a randomized open label study of immediate versus delayed treatment on 30 patients. At week 12, significant improvements were observed in self-perception, assessed by a visual analog scale; however, changes in anxiety and depression scores were not significant. Furthermore, no differences between the immediate and delayed group were observed at week 24, both in the anxiety and depression scores.

In contrast, a recent observational, prospective study⁴⁰ of 299 participants aimed in assessing long-term psychometric outcomes of surgical treatment showed significant improvements not only in facial aesthetic satisfaction, but also in body image satisfaction and in the depression measures.

Thus, despite all the published studies that consistently showed significant improvements in lipoatrophy grading, whether or not FLA filling interventions also resulted in improvements on HRQoL, depression and anxiety scores remain controversial.

Overall, the above-mentioned studies showed at least two kinds of limitations. The first problem is that PROs results evaluating the facial surgery are generally based on the data obtained in prospective, open-label, and uncontrolled studies, which introduce concerns related to placebo effects and limit the robustness of the results. The second limit is that none of these studies adopted lipoatrophy-specific instruments. To date, the ABCD questionnaire is the instrument more suitable than the other instruments mentioned to determine important psychosocial consequences of body changes, with particular regard to its implications for social and relational life. Nevertheless, the authors that validated the ABCD questionnaire²⁹ stated that a major limit of this instrument is the absence of any specific items referring to FLA.

To our knowledge, the present work was the largest randomized, controlled study published to date. In our study several PROs instruments were used and a large number of patients were studied. For these characteristics, results are expected to provide a significant contribution in order to evaluate properly the outcomes of filling surgical interventions. A possible restriction of the study is represented by the presence of patients with missing PROs information at the end of the study. As in the above studies, our research had an important limitation due to the absence of a specific lipoatrophy questionnaire. Thus, although results from all the PROs instruments we used showed no significant evidence that surgical interventions with facial fillers had a positive impact on the psychosocial well-being and health-related quality of life of HIV-infected persons with facial lipoatrophy, a robust conclusion cannot be drawn. Another potential limitation of our study is the absence of objective measures to assess changes in FLA severity. However, this does not diminish the significance of subjective improvements in FLA severity grade as determined by patients.

Moreover, a potential explanation for our negative findings on HRQoL is the complexity of factors associated with poor quality of life in HIV-infected patients with lipoatrophy.⁴¹ A consistent proportion of our patients suffered from several conditions associated with poor quality of life: a long history of HAART exposure, previous AIDS-related events, a considerable number of HAART-related side effects, and evidence of severe lipodystrophy involving other body sites (such as loss of fat in the legs and arms). Furthermore, it is plausible that treatment-experienced patients who have to face so many problems every day desire to have a normal lifestyle rather than a better physical appearance. Thus, a simple correction of FLA could have had a limited impact on overall HRQoL in patients suffering from a relevant number of factors affecting quality of life. Moreover, as far as anxiety is concerned, since anxiety is to a large extent a trait measure, the absence of large changes in anxiety scores was not an unexpected finding.

In conclusion, our study supports and supplements previous reports indicating relevant improvements in the FLA severity scale in patients with severe FLA treated with surgical interventions with facial fillers. However, no adequate conclusions can be drawn in terms of PROs evaluations, since at present there are no validated and reliable patient-centered instruments to evaluate correctly the full impact of lipoatrophy on psychosocial and HRQoL dimensions in HIV-infected people. In our opinion, efforts should be focused on the development of specific lipoatrophy instruments to address this specific issue. Furthermore, studies should be performed to identify patients who could maximally benefit from filling interventions for FLA.

Footnotes

Acknowledgments

This study was partially supported by Ministero della Sanità, Programma di Ricerca Corrente degli I.R.C.C.S. Pasquale Narciso and Raffaella Bucciardini contributed equally to this article.

Disclosure Statement

No competing financial interests exist.

References

Palella

, Delaney

, Moorman

et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med, 1998; 338:853–860.

, Lamping

. Assessment of quality of life in HIV disease. AIDS, 1994; 8:S349359.

Department of Health and Human Services (DHHS). Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. http://www.hivatis.org. 2008 October 14January 29, 2008.

Hengel

, Watts

, Lennox

. Benign symmetric lipomatosis associated with protease inhibitors. Lancet, 1997; 350:1596.

Roth

, Angel

, Kravcik

. Development of a cervical fat pad following treatment with HIV-1 protease inhibitors. 5th Conference on Retroviruses and Opportunistic Infections, Chicago, IL, February 1–5, 1998[Abstract 411].

Carr

, Samaras

, Burton

, Law

, Freund

, Chisholm

et al. A syndrome of peripheral lipodystrophy, hyperlipidemia, and insulin resistance in patients receiving HIV protease inhibitors. AIDS, 1998; 12:F51–F8.

, Mulligan

, Tai

, Algren

, Schambelan

. “Buffalo hump” in men with HIV-1 infection. Lancet, 1998; 351:867–870.

Schambelan

, Benson

, Carr

et al. Management of metabolic complications associated with antiretroviral therapy for HIV-1 infection: Recommendations of an International AIDS Society-USA Panel. J Acquir Immune Defic Syndr, 2002; 31:257–275.

Kotler

, Rosenbaum

, Wang

, Pierson

. Studies of body composition and fat distribution in HIV-infected and control subjects. J Acquir Immune Defic Syndr Hum Retrovirol, 1999; 353:2093–2099.

10.

Thiebaut

, Daucourt

, Mercie

et al. Lipodystrophy, metabolic disorders, and human immunodeficiency virus infection: Aquitaine Cohort, France, 1999. Clin Infect Dis, 2000; 31:1482–1487.

11.

Galli

, Polo

, Saint-Marc

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.

12.

Collins

, Wagner

, Walmsley

. Psycosocial impact of the lipodystrophy syndrome in HIV infection. AIDS Read, 2000; 10:546–550.

13.

Oette

, Juretzko

, Kroidi

et al. Lipodistrophy syndrome and self-assessment of well being and physical appearance in HIV-positive patients. AIDS Patient Care STDS, 2002; 16:413–417.

14.

Santos

, Felipe

, Braga

, Ramos

, Lima

, Segurado

. Self-perception of body changes in persons living with HIV/AIDS: Prevalence and associated factors. AIDS, 2005; 19:S14–S21.

15.

Sutinen

. Interventions for managing antiretroviral therapy-associated lipoatrophy. Curr Opin Infect Dis, 2005; 18:25–33.

16.

Lundgren

, Battegay

, Behrens

et al. European AIDS Clinical Society (EACS) guidelines on the prevention and management of metabolic diseases in HIV. HIV Med, 2008; 9:72–81.

17.

Tai

, Schambelan

, Algren

, Shayevich

, Mulligan

. Effects of recombinant human growth hormone on fat distribution in patients with human immunodeficiency virus-associated wasting. Clin Infect Dis, 2002; 35:1258–1262.

18.

, Mulligan

, Noor

, Schwarz

, Halvorsen

, Grunfeld

, Schambelan

. The effects of recombinant human growth hormone on body composition and glucose metabolism in HIV-infected patients with fat accumulation. J Clin Endocrinol Metab, 2001; 86:3480–3487.

19.

Ascher

, Coleman

, Alster

et al. Full scope of effect of facial lipoatrophy: A framework of disease understanding. Dermatol Surg, 2006; 32:1058–1069.

20.

Von Buelow

, Pallua

. Efficacy and safety of polyacrylamide hydrogel for facial soft-tissue augmentation in a 2-year follow-up: A prospective multicenter study for evaluation of safety and aesthetic results in 101 patients. Plast Reconstr Surg, 2006; 118:85S–91S.

21.

De Santis

, Jacob

, Baccarani

et al. Polyacrylamide hydrogel injection in the management of human immunodeficiency virus-related facial lipoatrophy: A 2-year clinical experience. Plast Reconstr Surg, 2008; 121:644–653.

22.

Moyle

, Lysakova

, Brown

et al. A randomized open-label study of immediate versus delayed polylactic acid injections for the cosmetic management of facial lipoatrophy in persons with HIV infection. HIV Med, 2004; 5:82–87.

23.

Cattelan

, Bauer

, Trevenzoli

et al. Use of polylactic acid implants to correct facial lipoatrophy in human immunodeficiency virus 1-positive individuals receiving combination antiretroviral therapy. Arch Dermatol, 2006; 142:362–364.

24.

Moyle

, Brown

, Lysakova

, Barton

. Long-term safety and efficacy of poly-L-lactic acid in the treatment of HIV-related facial lipoatrophy. HIV Med, 2006; 7:181–185.

25.

Bechara

, Gambichler

, Brockmeyer

, Sand

, Altmeyer

, Hoffmann

. Hyaluronic acid new formulation: Experience in HIV-associated facial lipoatrophy. Dermatology, 2008; 217:244–249.

26.

Carr

, Law

. An Objective lipodystrophy severity grading scale derived from the lipodystrophy case definition score. J Acquir Immune Defic Syndr, 2003; 33:571–576.

27.

Brooks

. the EuroQoL Group. EuroQoL: The current state of play. Health Policy, 1996; 37:53–72.

28.

Bucciardini

, Murri

, Guarinieri

et al. ISSQoL: A new questionnaire for evaluating the quality of life of people living with HIV in the HAART era. Qual Life Res, 2006; 15:377–390.

29.

Guaraldi

, Orlando

, Murri

et al. Quality of life and body image in the assessment of psychological impact of lipodystrophy: Validation of the Italian version of assessment of body change and distress questionnaire. Qual Life Res, 2006; 15:173–178.

30.

Conti

. Anxiety Status Inventory (ASI) and Self-Rating Anxiety Scale (SAS) Repertorio delle scale di valutazione in psichiatria, tomo secondo (Conti L, a cura di) Firenze, S.E.E., 1999; 537–543.

31.

Kolb

. Noyes' Modern Clinical Psychiatry, 7th edition. W.B. Saunders: Philadelphia, PA, 1968; 464–466.

32.

Lief

. Freedman

, Kaplan

. Anxiety reaction. Comprehensive Textbook of Psychiatry. Williams & Wilkins: Baltimore, MD, 1967; 865–866.

33.

Portnoy

. The anxiety states. American Handbook of Psychiatry. Arieti

Vol. I. Basic Books: New York, 1959; 308.

34.

Wheeler

, White

, Reed

, Cohen

. Neurocirculatory asthenia (anxiety neurosis, effort syndrome, neurasthenia) JAMA, 1950; 142:878–888.

35.

Zung

WWK

. A rating instrument for anxiety disorders. Psychosomatics, 1971; 12:371–379.

36.

Kazis

, Anderson

, Meenan

. Effect sizes for interpreting changes in health status. Med Care, 1989; 27,3 Suppl.:S178–189.

37.

Cohen

. Statistical Power Analysis for the Behavioral Sciences, 2nd ed.. Lawrence Erlbaum Associates, Inc.: Hillsdale, NJ, 1988.

38.

Loutfy

, Raboud

, Antoniou

et al. Immediate versus delayed polyacrylamide gel injections to correct facial lipoatrophy in HIV-positive patients. AIDS, 2007; 21:1147–1155.

39.

Lafaurie

, Dolivo

, Porcher

, Rudant

, Madelaine

, Molina

. Treatment of facial lipoatrophy with intradermal injections of polylactic acid in HIV-infected patients. J Acquir Immune Defic Syndr, 2005; 38:393–398.

40.

Orlando

, Guaraldi

, De Fazio

et al. Long-term psychometric outcomes of facial lipoatrophy therapy: Forty-eight-week observational, nonrandomized study. AIDS Patient Care STDS, 2007; 21:833–842.

41.

Blanch

, Rousaud

, Martinez

et al. Factors associated with severe impact of lipodystrophy on quality of life in patients infected with HIV. Clin Infect Dis, 2004; 38:1464–1470.