Correction of facial lipoatrophy in HIV-infected patients on highly active antiretroviral therapy by injection of autologous fatty tissue

Highly active antiretroviral therapy (HAART) has transformed the prognosis of HIV infection, but is associated with numerous adverse effects, including lipodystrophy syndromes [^1,2]. One of the most striking features of lipodystrophy is the loss of subcutaneous fat on the face, limbs and buttocks. Some patients find these changes in their body image very troublesome [³], sometimes leading to sexual dysfunction and poor adherence to treatment.

There is no medical treatment to correct this loss of fat. The benefit of switching these patients to other antiretroviral drug classes is controversial, but this strategy does not appear to improve facial atrophy in the mid term at least [⁴].

The surgical enhancement of facial contours has been practised for more than a century. Coleman's technique, or lipostructure, consists of harvesting, refining and re-injecting autologous fatty tissue [⁵]. We assessed the efficacy and safety of this technique in HIV-infected patients with facial lipoatrophy caused by HAART in an open prospective study.

All HIV-infected patients seeking treatment for stable (≥ 3 months) facial lipoatrophy were on HAART. Patients were excluded if they had wasting (> 10% weight loss compared with baseline), a CD4 T lymphocyte count below 100 cells/mm³, contraindications to general anaesthesia, or inadequate abdominal subcutaneous fat.

Abdominal subcutaneous adipose tissue was harvested with a blunt cannula and centrifuged for 3 min at 3000 rpm to separate blood and fat. The specific facial morphology of each patient was carefully analysed before re-injecting the purified fat. Patients were hospitalized for 2 days and given anti-staphylococcal prophylaxis.

Patients were seen 7 and 28 days and 1, 3 and 6 months after the operation. A self-administered satisfaction questionnaire was completed at 6 months. The questionnaire assessed the psychological impact of the facial lipoatrophy and its correction, together with the degree of improvement, facial symmetry after the procedure, and the global results. In addition, a five-member jury rated pictures of the patients taken at baseline and 3 and 6 months after the procedure. The pictures were all taken in standardized conditions. The rating focused on the correction of atrophy, symmetry, the global result and changes between month 3 and month 6. Transverse facial magnetic resonance imaging (MRI) was performed by the same radiologist at baseline and 6 months after the procedure in order to measure facial subcutaneous fat thickness at the level of the dental apices.

Fifteen patients (12 men, three women) were treated over a 6-month period. Their mean age was 44 years and their mean body mass index 23.5. The mean interval between the diagnosis of HIV infection and lipostructure was 117 months (50-171) and the mean duration of antiretroviral therapy was 60.3 months (24-112). All the patients were receiving or had received protease inhibitors.

The main adverse effects of lipostructure were moderate swelling at the injection sites (six patients) and moderate pain at the donor sites (five patients). Manual palpation was normal in all cases and the consistency of the skin appeared normal.

The results of the questionnaire-based self-assessment at month 6 and the results of the jury evaluation are shown in Table 1. Thirteen of the 15 patients scored the atrophy correction 'good' or 'very good' and the symmetry 'slightly asymmetric' or 'symmetric'. Fourteen patients scored the global results 'acceptable', 'good' or 'very good'.

Table 1

Thirteen of the 14 patients evaluated by the jury were considered to have 'acceptable', 'good' or 'very good' global results at month 6. The correction of facial atrophy and the global results improved from month 3 to month 6 in four out of 14 (29%) and eight out of 14 patients (57%).

MRI showed a mean increase in fat thickness between baseline and month 6 of 10.5 mm (3-22) and 10 mm (3-21) on the left and right side, respectively. Photographic and MRI evaluation of one patient is shown in Fig. 1. Only one patient had an antiretroviral treatment modification during follow-up (zidovudine replacing stavudine).

Fig. 1

Lipostructuring, the re-injection of autologous fatty tissue, gave very acceptable results in this series of patients with HAART-associated facial lipoatrophy. Six months after the procedure, 74% of the patients were highly satisfied with the results, which were considered by an assessment panel to be 'very good' in 94% of cases. MRI confirmed a persistent increase in facial fat thickness at 6 months.

One important advantage of this technique is the lack of allergic reactions observed with most injected filling substances. The injected adipocytes rapidly integrate their new environment, yielding a durable correction of facial atrophy. The long-term follow-up (6.5 years) of fat transplants in HIV-uninfected patients shows that fatty tissue harvested and handled with care and re-injected with the technique used here remain stable [⁶]. Interestingly, our results were considered better at 6 months than at 3 months, whereas the efficacy of non-living materials such as collagen and hyaluronic acid is transient because of rapid degradation.

This technique is simple, low risk and yields a high rate of patient satisfaction. The requirement for a sufficient amount of abdominal or lumbar subcutaneous fat is the main limitation of this procedure. A longer follow-up is needed to ensure that the correction persists.