- Email: [email protected]
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy using three-dimensional surface laser scanning and psychological assessment
Journal of Plastic, Reconstructive & Aesthetic Surgery (2009) 62, 1627e1635
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy using three-dimensional surface laser scanning and psychological assessment J. Ong a,*, A. Clarke a, P. White b, M.A. Johnson c, S. Withey a, P.E.M. Butler a a
Department of Plastic and Reconstructive Surgery, The Royal Free Hospital Hampstead NHS Trust, Pond Street, London, NW3 2QG, UK b Department of Statistics, University of the West of England, Frenchay Campus, Coldharbour Lane, Bristol, BS16 1QY, UK c Department of Infectious Diseases and HIV Medicine, The Royal Free Hospital Hampstead NHS Trust, Pond Street, London, NW3 2QG, UK Received 19 March 2008; accepted 17 July 2008
KEYWORDS Disfigurement; Facial lipoatrophy; HIV; Laser scanning; PLA; Psychology
Summary The advent of highly active antiretroviral therapy (HAART) has dramatically improved the life expectancy of people infected with human immunodeficiency virus (HIV). Although patients often have excellent disease control with these combinations of antiretrovirals, they are at risk for the multiple toxicities associated with these drugs. Facial lipoatrophy is a particularly distressing complication of some HAART regimes. This disfigurement can lead to significant psychosocial stress, resulting in decreased treatment compliance. Polylactic acid (PLA) facial implants provide a potential method of restoring a normal appearance. One hundred consecutive patients had a course of PLA facial implants. All patients were assessed clinically and had photographs, facial surface laser scans and completed psychological questionnaires throughout the course of treatment. After a mean of 4.85 treatments per patient, there were improvements in all measures. The mean clinical scores improved from a moderateesevere grade to noneemild grade after treatment. Threedimensional (3D) laser surface scans showed a volume increase of 2.81 cc over the treated area of the cheek. There were significant improvements in all of the psychological measures. This study shows clear objective evidence of the psychological and physical benefit of PLA implants in HIV-associated facial lipodystrophy. ª 2008 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Address: Department of Plastic and Reconstructive Surgery, The Royal Free Hospital Hampstead NHS Trust, Pond Street, London NW3 2QG, UK. Tel.: þ44 7903 858 297. E-mail address: [email protected] (J. Ong). 1748-6815/$ - see front matter ª 2008 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2008.07.046
1628 The morbidity and mortality of human immunodeficiency virus (HIV) infection has been significantly improved with the introduction of highly active antiretroviral therapy (HAART).1 However, patients on these treatments are at risk for complications and toxicities of the treatment regime itself.2e4 The best medical option to prevent progression of lipodystrophy would be to stop the associated antiretroviral combination and switch to an alternative regime. These complications of HAART may present individually as metabolic disturbances or discrete morphological changes or together as part of the lipodystrophy syndrome. Patients may develop craniofacial lipoatrophy, limb lipoatrophy, central abdominal obesity, gynaecomastia, ‘double chin’ and ‘buffalo hump’. HIV-associated facial lipoatrophy particularly affects the buccal and malar fat pads, resulting in a characteristic pattern of facial disfigurement. The wasting leaves a volume deficit in the subcutaneous compartment which causes the skin to sag into depressions defined by the underlying bony and muscular structure. The increased visibility of the underlying structures results in a ‘sick’ or emaciated look in patients who often have optimised disease control. These physical changes in body shape, particularly of the face, can contribute to significant psychological morbidity in patients who are otherwise well on successful HAART regimes.5 Apart from the stigma associated with the physical disfigurement, these characteristic changes can potentially identify the HIV status of the individual to others. This loss of medical anonymity amongst an oftenstigmatised population may contribute to an increased incidence of psychological illness. Patients often adopt coping measures similar to other patients with facial disfigurement such as disguising affected areas with hairstyles or clothes as well as avoiding social interaction. Importantly, some patients choose to stop their antiretroviral treatment.6 Current strategies for the management of facial lipoatrophy must include altering HAART regimes associated with lipodystrophy. Treatment options range from local flaps and autologous tissue-transfer techniques to correcting the deficit using biological or synthetic fillers which are injected locally.7 Whilst there has been a variable degree of success with different forms of fillers in terms of complications and permanence of change, polylactic acid (PLA) provides a promising medium-term treatment. A good cosmetic result for patients has the aim of normalising appearance, reducing fears of stigma and reducing psychological morbidity. The synthetic biodegradable implant PLA (Sculptra) is made of small (approx. 50-mm diameter) particles. The PLA has a long history of use in humans in the form of sutures, orthopaedic implants and more recently in sustainedrelease drug preparations. It is immunologically well tolerated and hydrolyses to lactic acid which is metabolised and ultimately excreted as CO2. These implants are injected as a suspension via a small-bore needle into the subdermal tissues. The particles act as a stimulus for fibroblast recruitment and as scaffold for the deposition of neo-collagen by these fibroblasts.8 The resulting filling effect is, therefore, independent of the lipodystrophic process.
J. Ong et al. Facial wasting in HIV-associated lipoatrophy often has an insidious onset and can go unnoticed by both patient and physician until a relatively advanced stage. The contours of the face are extremely complex and variable between individuals. Monitoring subtle volume changes over long time periods in an ageing population is particularly difficult. Facial laser surface scanning is repeatable and accurate9 and has been reported to have a higher sensitivity in measuring the subtle changes of HIV-associated facial lipoatrophy than clinical examination10 or facial anthropometry.11
Patients and methods Design This was a prospective study to objectively measure the efficacy of PLA in correcting the physical features of HIVassociated facial lipoatrophy and the effects of an aesthetic correction on psychological morbidity associated with facial disfigurement.
Sample One hundred consecutive patients with HIV-associated facial lipoatrophy were recruited from the Infectious Diseases Department at the Royal Free Hospital. All the patients were over the age of 18 years and had lipoatrophy grade confirmed by clinical examination by two doctors within the department (one doctor was usually the referring infectious diseases physician, the other being the plastic surgeon performing the treatment). Complete data are available on 65 patients. All patients were male, Caucasian, had no previous history of facial implants and had only had a single course of treatment. These treatments consisted of multiple small-volume (approximately 100 ml) injections of a suspension of PLA into the subcutaneous tissues of each cheek. The volume of suspension injected was adjusted according to severity of the lipoatrophy and to compensate for any asymmetry. Each vial was reconstituted to a volume of 5 ml, with 1.5 ml of lidocaine (2%) and 3.5 ml of water for injection. Two vials of PLA (Sculptra) were injected into each patient at each session. A firm facial massage was performed over the treatment area immediately after treatment. All patients were advised to massage their cheeks at regular intervals for 5 days. The patients underwent a total of four or five treatments at 4-week intervals. At each appointment, patients were assessed for any complications of treatment (e.g., bruising, infection and asymmetry). All the patients were also clinically reviewed 1 month following the last treatment and to ensure that all assessment measures had been completed and that followup had been confirmed. Ethical approval was granted by the local research ethics committee.
Measurement tools The patients were examined clinically using the regional scoring system used in the HIV Outpatient Study12 (HOPS)
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy and photographed. In this system, a score of 0 (none) represents no clinical facial lipoatrophy; 1, (mild) lipoatrophy detectable on close examination; 2, (moderate) lipoatrophy easily visibly to patient and doctor and 3, (severe) facial lipoatrophy obvious to a casual observer. These were recorded at 6-monthly intervals (pre-treatment, 6, 12, 18 and 24 months). Assessment of complex surfaces, such as faces, is difficult to perform accurately and objectively. We used 3D surface laser scanners as they are sensitive, quick to use and are comparable over time. By using the patient’s face as the baseline, we could provide an individualised internal control and thereby have a process which would allow an accurate detection of subtle changes in face surface. The 3D laser scans were performed on the same machine (University College London (UCL) optical face scanner) by the Maxillofacial Department of University College London. The patients were scanned pre-treatment and at 6-monthly intervals for 2 years during the follow-up period. These scans were then stored for comparison and analysis using Prima Facie software. The images were matched using fixed non-changing, untreated areas (e.g., forehead, ears, nose and chin). All scans were compared to the pre-treatment scan, and the volume changes were measured. Psychological factors associated with altered appearance were measured using the Hospital Anxiety and Depression Scale (HADS), the Rosenberg Self-Esteem Scale (RSE) and the Derriford Appearance Scale (DAS59). The HADS scale13 is a 14-item screening tool for anxiety and depression, standardised on both general and hospital populations. The HADS uses cut-off points as follows: normal (0e7), mild (8e10), moderate (11e15) and severe clinical range (16e21). The RSE14 is a short (10-item) scale, where low scores indicate high self-esteem. Normative data in the original test sample is skewed towards low self-esteem, but the instrument is sensitive to change and has been widely reported in the field of clinical psychology. The DAS5915 is a 59-item factorial scale measuring appearance-related distress, social anxiety and avoidance, standardised on both general and hospital populations. The DAS59 comprises subscales examining general selfconsciousness of appearance (GSC), social self-consciousness (SSC), self-consciousness of body and sexual selfconsciousness (SBSC), negative self-consciousness (NSC), self-consciousness of facial appearance (FSC), physical distress and dysfunction (PDD) as well as a whole scale (WS) score. Higher scoring on each of these subscales is associated with increasing psychosocial morbidity. The psychological measures of well-being (the Derriford subscales, subscales of HADS and self-esteem) were recorded at pretreatment, 6, 12, 18 and 24 months for each participant. All of the psychometric scales used in this study have good psychometric and scale properties. The multiple measures repeatedly measured on each participant may be analysed as a one-way, doubly multivariate, repeatedmeasures analysis of variance (ANOVA),16 with follow-up one-way, repeated-measures ANOVA for each measure. Application of Mauchly’s test of sphericity indicates that there is no basis for rejecting the sphericity hypothesis for the data collected and indicates that the analysis is valid. A pair-wise post hoc application of the paired-samples t-test
1629
on each measure, with Bonferroni corrected levels of observed significance, was used to locate precise differences in each measure over the five time points. The changes examined did not suggest gross deviations from normality, and, given the sample sizes in this study, the use of the paired-samples t-test is defendable as having robust validity and efficiency.17 The degree of the strength of the relationship between laser scores and each measure of well-being can be quantified using Spearman’s correlation coefficient. The laser scores at each 6-month interval represent a change from the baseline, and, for this reason, the strength of the relationship between laser scores and the corresponding change in each measure of well-being is also quantified using Spearman’s correlation coefficient.
Results Demographics All patients in this study were male and Caucasian. The mean age was 45 years, with a range of 30e72 years. The duration of infection with HIV was between 2 and 23 years (mean: 11.1 years). Of these patients, 46% had a clinical stage of acquired immunodeficiency syndrome (AIDS), 44% had symptomatic HIV infection and 10% had asymptomatic infection. The following analyses are based on a sample of n Z 65 patients who have provided data at each time interval. A missing-value analysis has been performed comparing baseline psychological measures between the group of patients with complete data (n Z 65) and those with incomplete data (n Z 35). These analyses indicate that there are no systematic differences between the two groups.
Clinical examination Mean values and standard deviations of clinical scores are included in Table 1. The distribution of severity of clinical lipoatrophy is illustrated in Figure 1, showing that clinical grades were either moderate or severe prior to treatment (mean: 2.82). These grades dropped to 0.66 (noneemild) after treatment (6 months) and progressively increased again to 1.26 at the 24-month review.
Clinical photography All patients had antero-posterior (AP), left- and rightlateral views of the face performed at 6-monthly intervals. These photographs were used as objective records of the scores given by clinical examination. Figure 2 shows preand post-treatment photographs of the same patient illustrated in the 3D laser scans.
3D laser scans Laser scans were performed before each treatment and before each subsequent 6-monthly review. Figure 3 shows computer-generated images of the laser scans. When the baseline scans (Figure 3, Left) are compared to posttreatment images (Figure 3, Right), a composite image is generated (Figure 3, Centre). The composite image shows
1630
J. Ong et al.
Table 1
Mean values and standard deviations of all measures at each time point
Measures of self-consciousness of appearance
Derriford Appearance Scale
Hospital Anxiety and Depression Scale Rosenberg Self Esteem Scale Change in volume over treated area (c.c.) Clinical Grade
Whole Score General self-consciousness Social self-consciousness Body/sexual self-consciousness Negative self-concept Facial self-consciousness Physical distress and dysfunction Anxiety Depression
Time (in months) Baseline
6
122.42 29.20 35.63 17.83
94.23 19.66 24.37 12.18
(35.10) (12.60) (13.93) (9.25)
15.23 (4.82) 5.09 (3.97) 2.48 (1.26)
16.34 (4.82) 4.12 (2.97) 2.49 (1.48)
10.25 (4.34) 7.51 (4.46)
7.23 (3.94) 4.00 (3.54)
8.68 (4.32) 5.35 (3.75)
9.02 (4.52) 5.05 (3.80)
10.26 (4.66) 5.28 (3.53)
22.05 (6.54)
18.45 (5.26)
19.97 (4.68)
20.55 (4.11)
21.94 (5.59)
2.81 (0.92)
3.00 (1.00)
2.99 (1.08)
2.68 (1.06)
0.66 (0.54)
0.87 (0.64)
1.06 (0.46)
1.26 (0.56)
Severe
70 2
60
10
9 20
Frequency
50 40
38 56
39 51
30
41 20 25
10
16
9 0 Baseline
6 Months
12 Months
(25.99) 109.68 (35.90) 113.46 (32.52) (8.76) 22.00 (8.68) 26.86 (9.65) (9.25) 29.14 (11.43) 29.12 (8.74) (7.05) 13.86 (7.89) 16.75 (8.83)
15.06 (4.08) 3.71 (2.67) 2.54 (1.43)
the differences between the surfaces of the baseline and the subsequent scan as a colour change. The scans from each time point were compared to the baseline (pre-treatment) scan to calculate volume changes over the treated area. The mean volume change from baseline to each time point is shown in Figure 4. The onesample t-test (scores at the baseline are all equal to 0) comparing all time points to the baseline shows that, after treatment the mean cheek volumes (per cheek) were all significantly different from 0 (mean difference Z 2.81 cc, p < 0.005). Differences in measurements of cheek volume compared to the baseline measurements and those between 6, 12, 18 and 24 months were assessed using the one-way repeated-measures ANOVA. A pair-wise application of the paired-samples t-test with Bonferroni corrected levels of significance confirmed that there were no significant differences in mean cheek
Moderate
98.66 20.31 26.37 14.17
24
13.52 (4.76) 3.22 (2.69) 2.69 (1.61)
2.82 (0.38)
Mild
(28.71) (7.48) (8.58) (7.95)
18
16.71 (5.05) 6.85 (7.79) 3.65 (2.17)
———————————
None
12
5
4
18 Months
24 Months
Time (Months)
Figure 1 The frequency distributions of clinical lipoatrophy scores over time.
projections between 6-, 12- and 18-month time points. However, a comparison of mean results between 18- and 24-month periods show a small, but statistically significant, decrease in mean cheek projection (p < 0.001).
Psychological measures Table 1 summarises the sample means (and standard deviations) for each measure for the five time points. Application of a one-way, doubly multivariate, repeated-measures ANOVA with the subscales of the DAS59 and measures of anxiety (HADS), depression (HADS) and self-esteem (RSE) as the dependent variables indicates that there are multivariate, statistically significant changes over the 24-month period (Wilks’ lambda Z 0.149, multivariate F(36, 29) Z 3.73, p < 0.001). Application of separate, one-way, repeated-measures ANOVA’s for each measure of well-being indicate that there is a statistically significant change in each subscale of DAS59, and in anxiety, depression and self-esteem over the five time points (p < 0.001 in all cases). A post hoc pair-wise application of the paired-samples t-test with Bonferroni corrected levels of significance for two-sided tests has been carried out for each measure. The Bonferroni adjusted p-values for comparisons at 6, 12, 18 and 24 months when compared with the baseline measures are summarised in Table 2. All measures of wellbeing show a statistically significant improvement after treatment at 6-month follow-up when compared with the baseline, except for physical distress (p Z 0.131). At the 24-month review, all measures of well-being, except for social self-consciousness (p Z 0.002), physical distress (p Z 0.004) and depression (p Z 0.028), returned to the baseline (pre-treatment) values. The pattern of statistical significance is particular to the measure under investigation (Table 2).
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy
Figure 2
1631
Clinical photographs before (left) and after (right) treatment.
Correlations between measures Laser scores (facial-volume changes) represent a change from the baseline scores. Accordingly, changes from the baseline in the measures of well-being were correlated with laser scores. At 6 months, only the change in facial self-consciousness was significantly correlated with the laser score (r Z 0.246, n Z 65 and p Z 0.048, two-sided). At 12 months, none of the changes in well-being was significantly correlated with the laser scores. At 18 months, the change in scores for negative self-consciousness and the laser scores did show a statistically significant correlation (r Z 0.313, n Z 65 and p Z 0.011, two-sided) but all other correlations were not significant. At 24 months too, none of the changes in well-being was significantly correlated with laser scores. Between the baseline and the 6-month time point, all participants showed an improvement in clinical scores coinciding with a change in the laser scores. Out of the 65 participants, 52 (i.e., 80% of the participants) showed an improvement on six or more measures of well-being. At 6 months, 25 participants had a clinical score of 0,
representing no clinical facial lipoatrophy. However, a multivariate ANOVA comparing those without clinical facial lipoatrophy with those showing some signs of facial lipoatrophy indicates that there is no significant difference on the measures of well-being between the two groups (Wilks’ Lambda Z 0.903, Multivariate F(9, 55) Z 0.583, p Z 0.821). For 27 patients, there were no changes in clinical grading between 6 and 24 months, whereas 35 patients showed a worsening in clinical grading. A multivariate analysis of the measures of well-being showed that the changes in measures of well-being between 6 and 24 months were not significantly different between those who did not have a change in clinical grading and those whose clinical scoring had increased (Wilks’ Lambda Z 0.778, Multivariate F(9, 52) Z 1.652, p Z 0.125). At 24 months, 20 patients had a clinical score of 2 (i.e., (moderate) lipoatrophy easily visible to patient), and the remaining 45 patients either had no clinically apparent facial lipoatrophy or had mild facial lipoatrophy that was only detectable on close examination. A multivariate ANOVA indicates that those cases showing moderate lipoatrophy that is visible to the patient at 24 months do not differ on the measures of
Figure 3 Laser-scan images showing the right cheek before (left) and after (right) treatment. The composite image (centre) shows the difference in scans represented by colour. Scale in mm is depicted at the bottom of the image. Note positive projection of skin surface of cheek (up to 8 mm) and negative change in beard area from shaving.
1632
J. Ong et al.
Figure 4 Ninety-five percent confidence intervals for change in cheek volume (cc) measured using laser surface scanning.
well-being from those cases with mild or no facial lipoatrophy (Wilks’ Lambda Z 0.902, Multivariate F(9, 55) Z 0.662, p Z 0.739).
Complications At each appointment, patients were asked specifically about complications from their previous treatment. Bruising was the most common complication, being reported after 7.5% of treatments. The only other reported complication was subcutaneous lumps. These occurred in 3.3% of patients, and all the lumps had resolved at the 1-year follow-up. All complications were treated conservatively.
Discussion Treating facial wasting with PLA implants is an effective way to normalise the disfigurement associated with HIVassociated facial lipoatrophy. In this study, patients had moderate-to-severe facial lipoatrophy, and treatment
reduced the clinical grade to a none-to-mild grade after four or five treatments. Previous studies have shown that psychosocial distress is related to physical deformity and facial disfigurement,18,19 although there is no evidence for a direct correlation between severity of disfiguring conditions and associated psychological distress. Facial lipoatrophy is not only disfiguring, but can also identify the HIV status of the individual in knowledgeable communities. The involuntary disclosure of HIV status can impact adversely on the well-being as a result of stigmatisation, discrimination and disruption of personal relationships.19,20 Treatment of HIV-associated facial lipoatrophy provides a naturally occurring AeBeA21 design with repeated measures taken before treatment (A), during treatment (B) and when conditions return to baseline (A). The treatment with PLA is successful in reducing the disfigurement. In addition, as the implant degrades (PLA is a synthetic polymer which hydrolyses and degrades by hydrolysis over time), we can examine the psychological impact of partially re-acquiring the deformity prior to starting a second course of treatment. Furthermore, facial wasting has little, if any, effect on normal physiology and any influence on social functioning is a reflection of the aesthetic deformity and its associated stigma. We found that mean 12-month and 18month cheek volumes are higher than the mean posttreatment (6-month) scans, thus showing continuing volume increase even after the end of treatment. This is illustrated in Figure 5, where improvements in cheek profile continue to progress after the end of the treatment process. Furthermore, although laser scans detected a significant drop in cheek volume from 18 to 24 months, the cheek volumes at 24 months were similar to the volumes immediately after treatment and significantly higher than baseline (pre-treatment) volume measurements. This can be explained by the relatively slow process of new collagenous tissue generation in response to the PLA implants, rather than direct volume increase due to the implants alone. (Patients generally only report subjective improvements after the third or fourth treatment.) Comparisons of the time points show that even at 24 months (although there is some volume loss from the peak at 12 months), there is still no statistically significant difference in mean volumes from those measured in the 6-
Table 2 Bonferroni corrected p-values from a pair-wise application of the paired samples t-test for each psychological measure between each time point and baseline Measure
Derriford Appearance Scale
HADS Rosenberg Self Esteem Scale
Time
Whole score General self-consciousness Social self-consciousness Body/sexual self-consciousness Negative self-concept Facial self-consciousness Physical distress and dysfunction Anxiety Depression
6
12
18
24
<0.001 <0.001 <0.001 0.002 0.001 0.013 0.131 <0.001 <0.001 <0.001
<0.001 <0.001 <0.001 0.167 0.538 0.033 0.010 0.576 0.050 0.462
0.252 0.004 0.022 0.094 0.925 1.000 0.002 1.000 0.015 1.000
1.000 1.000 0.002 1.000 1.000 0.123 0.004 1.000 0.028 1.000
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy
1633
Figure 5 Sequential clinical photographs illustrating facial changes; pre-treatment (left), post-treatment (centre) and at 12 months post treatment (right).
1634 month scans. This contrasts with a 0.6-point increase in clinical grades by the examining doctors between these two review points. Since it is unlikely that clinical judgement is more sensitive than the laser scanner in detecting a return in facial wasting, these data suggest a clinical bias towards placing a patient who may be of borderline grade in a classification of a worse clinical grade, particularly if the patient himself or herself perceives a change. Although the pattern of change was broadly similar between the different physical measures (laser and clinical), there were interesting differences in the psychological scales. The psychological questionnaires measuring appearance-related anxiety, anxiety, depression and selfesteem all showed benefits of treatment of facial lipoatrophy with PLA implants. Most of the scales had a similar pattern of change, namely a decrease in psychological morbidity after treatment, with a decrease in this improvement over time, but there were differences in the behaviour of individual scale scores during the review phase. The GSC, SSC, FSC and PDD had the most durable benefit, with mean values not returning to baseline levels at the 24month review point. Similarly, the improvements in the depression subscale of HADS also lasted up to 24 months. However, anxiety (HADS) scores fell to subclinical levels at 6 months, but reverted back to the baseline level at 24 months, moving back into the moderate clinical range. The SBSC and NSC returned to baseline levels at 24 months after treatment. The FSC levels dropped after treatment just as the levels in other subgroups of DAS did after treatment, but increased to levels reflecting higher social anxiety than the 24-month scores between the 12and 18-month reviews. At this point, there are no objective physical changes measured by laser scan measures, and mean clinical grades are still significantly lower than baseline grades. These scores, together with the elevation of the anxiety subscale of the HADS, support an explanation in terms of anticipatory anxiety. The patients are aware that PLA implants are not permanent, and the anticipation of a return in symptoms impacts on psychological wellbeing ahead of physical change. Anxiety is also likely to increase vigilance and scrutiny of the face, with patients who worry about the return of the disfigurement and its impact on their lifestyle, perceiving the onset of deterioration prior to its objective recurrence. This explanation is consistent with the observed bias in clinical grading for borderline cases. Analysis of laser-scan changes and psychological scores before and after treatment showed no correlation between the two variables in any of the psychological scales. This lack of correlation between the degree of physical deformity and level of psychological distress is consistent with the reports in literature and suggests that psychosocial adjustment is not directly related to the degree of deformity, but is modulated by individual factors such as the value placed on appearance, the subjective rating of visibility to others and the perceived likelihood of change. This study showed that treatment with PLA has a low incidence of complications. Bruising occurred after 7.5% of treatments. This was unpredictable and all patients had normal haematological parameters. All bruises settled spontaneously within 1e2 weeks. The most serious complication was the formation of temporary lumps. When
J. Ong et al. present, these were usually palpable at the post-treatment review and at the 6-month review. All had resolved by the time patients were reviewed 1 year after treatment. PLA implants improve the physical changes of HIV-associated facial lipoatrophy. They are easy to administer and produce a gradual improvement in facial profile over time. The absence of a sudden change also provides patients the opportunity to disguise the fact that they are having facial treatment. The physical changes shown by 3D surface laser are significant and last up to 24 months following treatment. However, the anticipation of a return in facial wasting at the end of treatment leads to a fall in psychological well-being when treatment ends despite negligible physical change measured on the cheeks. The clinical grading is more sensitive than laser scanning in predicting this change in psychological functioning. This is because it is driven by the objective assessment by the clinician as well as the subjective opinion of the patient and, hence, is more likely to be sensitive to any suggestion of change. Because of the impact of anticipated change, it is important to provide an explanation of the mechanism of collagen infiltration and reassurance that physical change will not be contiguous with treatment cessation. Re-assurance that treatment can be repeated to maintain appearance within the normal range will also be beneficial.
Acknowledgements The authors are very grateful for the kind help of all the patients and staff of the Plastic Surgery and Infectious Diseases Departments at the Royal Free Hospital who were involved with the clinical aspect of this study. Special thanks are also due to the medical imaging department at University College Hospital for their expertise in the use of the laser scanning equipment and help with the analysis of the images.
References 1. Palella Jr FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998;338:853e60. 2. Carr A. HIV lipodystrophy: risk factors, pathogenesis, diagnosis and management. Aids 2003;17:S141e8. 3. Carr A, Samaras K, Chisholm DJ, et al. Pathogenesis of HIV-1-protease inhibitor-associated peripheral lipodystrophy, hyperlipidaemia, and insulin resistance. The Lancet 1998;351: 1881e3. 4. Lichtenstein KA, Delaney KM, Armon C, et al. Incidence of and risk factors for lipoatrophy (abnormal fat loss) in ambulatory HIV-1-infected patients. J Acquir Immune Defic Syndr 2003;32: 48e56. 5. Barreiro P, Garcia-Benayas T, Soriano V, et al. Simplification of antiretroviral treatmentehow to sustain success, reduce toxicity and ensure adherence avoiding PI use 1. AIDS Rev 2002; 4:233e41. 6. Spire B, Duran S, Souville M, et al. Adherence to highly active antiretroviral therapies (HAART) in HIV-infected patients: from a predictive to a dynamic approach. Soc Sci Med 2002;54:1481e96. 7. Nelson L, Stewart KJ. Plastic surgical options for HIV-associated lipodystrophy. J Plast Reconstr Aesthet Surg 2008;61: 359e65.
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy 8. Gogolewski S, Jovanovic M, Perren SM, et al. Tissue response and in vivo degradation of selected polyhydroxyacids: polylactides (PLA), poly(3-hydroxybutyrate) (PHB), and poly(3hydroxybutyrate-co-3-hydroxyvalerate) (PHB/VA). J Biomed Mater Res 1993;27:1135e48. 9. Bush K, Antonyshyn O. Three-dimensional facial anthropometry using a laser surface scanner: validation of the technique. Plast Reconstr Surg 1996;98:226e35. 10. Benn P, Ruff C, Cartledge J, et al. Overcoming subjectivity in assessing facial lipoatrophy: is there a role for three-dimensional laser scans? HIV Med 2003;4:325e31. 11. Aung SC, Ngim RCK, Lee ST. Evaluation of the laser scanner as a surface measuring tool and its accuracy compared with direct facial anthropometric measurements. Br J Plast Surg 1995;48: 551e8. 12. Lichtenstein KA, Ward DJ, Moorman AC, et al. Clinical assessment of HIV-associated lipodystrophy in an ambulatory population. Aids 2001;15:1389e98. 13. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361e70. 14. Rosenberg M. Society and the adolescent self-image. Wesleyan University Press; 1989. 15. Harris DL, Carr AT. The Derriford Appearance Scale (DAS59): a new psychometric scale for the evaluation of patients with
16.
17. 18.
19.
20.
21.
1635
disfigurements and aesthetic problems of appearance. Br J Plast Surg 2001;54:216e22. Chaffin WW, Rhiel SG. The effect of skewness and kurtosis on the one-sample T test and the impact of knowledge of the population standard deviation. J Stat Comput Simulation 1993; 46:79e90. Kirk R. Procedures for the behavioural sciences. International Thompson Publishing; 1995. Stricker G, Clifford E, Cohen LK, et al. Psychosocial aspects of craniofacial disfigurement. A ‘‘State of the Art’’ assessment conducted by the Craniofacial Anomalies Program Branch, The National Institute of Dental Research. Am J Orthod 1979;76: 410e22. Harcourt DM, Rumsey NJ, Ambler NR, et al. The psychological effect of mastectomy with or without breast reconstruction: a prospective, multicenter study. Plast Reconstr Surg 2003; 111:1060e8. Chandra PS, Deepthivarma S, Jairam KR, et al. Relationship of psychological morbidity and quality of life to illness-related disclosure among HIV-infected persons. J Psychosom Res 2003; 54:199e203. Rosenthal R, Rosnow RL. Essentials of behavioural research: methods and data analysis. 3rd ed. McGraw Hill Higher Education; 2008.
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy using three-dimensional surface laser scanning and psychological assessment J. Ong a,*, A. Clarke a, P. White b, M.A. Johnson c, S. Withey a, P.E.M. Butler a a
Department of Plastic and Reconstructive Surgery, The Royal Free Hospital Hampstead NHS Trust, Pond Street, London, NW3 2QG, UK b Department of Statistics, University of the West of England, Frenchay Campus, Coldharbour Lane, Bristol, BS16 1QY, UK c Department of Infectious Diseases and HIV Medicine, The Royal Free Hospital Hampstead NHS Trust, Pond Street, London, NW3 2QG, UK Received 19 March 2008; accepted 17 July 2008
KEYWORDS Disfigurement; Facial lipoatrophy; HIV; Laser scanning; PLA; Psychology
Summary The advent of highly active antiretroviral therapy (HAART) has dramatically improved the life expectancy of people infected with human immunodeficiency virus (HIV). Although patients often have excellent disease control with these combinations of antiretrovirals, they are at risk for the multiple toxicities associated with these drugs. Facial lipoatrophy is a particularly distressing complication of some HAART regimes. This disfigurement can lead to significant psychosocial stress, resulting in decreased treatment compliance. Polylactic acid (PLA) facial implants provide a potential method of restoring a normal appearance. One hundred consecutive patients had a course of PLA facial implants. All patients were assessed clinically and had photographs, facial surface laser scans and completed psychological questionnaires throughout the course of treatment. After a mean of 4.85 treatments per patient, there were improvements in all measures. The mean clinical scores improved from a moderateesevere grade to noneemild grade after treatment. Threedimensional (3D) laser surface scans showed a volume increase of 2.81 cc over the treated area of the cheek. There were significant improvements in all of the psychological measures. This study shows clear objective evidence of the psychological and physical benefit of PLA implants in HIV-associated facial lipodystrophy. ª 2008 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Address: Department of Plastic and Reconstructive Surgery, The Royal Free Hospital Hampstead NHS Trust, Pond Street, London NW3 2QG, UK. Tel.: þ44 7903 858 297. E-mail address: [email protected] (J. Ong). 1748-6815/$ - see front matter ª 2008 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2008.07.046
1628 The morbidity and mortality of human immunodeficiency virus (HIV) infection has been significantly improved with the introduction of highly active antiretroviral therapy (HAART).1 However, patients on these treatments are at risk for complications and toxicities of the treatment regime itself.2e4 The best medical option to prevent progression of lipodystrophy would be to stop the associated antiretroviral combination and switch to an alternative regime. These complications of HAART may present individually as metabolic disturbances or discrete morphological changes or together as part of the lipodystrophy syndrome. Patients may develop craniofacial lipoatrophy, limb lipoatrophy, central abdominal obesity, gynaecomastia, ‘double chin’ and ‘buffalo hump’. HIV-associated facial lipoatrophy particularly affects the buccal and malar fat pads, resulting in a characteristic pattern of facial disfigurement. The wasting leaves a volume deficit in the subcutaneous compartment which causes the skin to sag into depressions defined by the underlying bony and muscular structure. The increased visibility of the underlying structures results in a ‘sick’ or emaciated look in patients who often have optimised disease control. These physical changes in body shape, particularly of the face, can contribute to significant psychological morbidity in patients who are otherwise well on successful HAART regimes.5 Apart from the stigma associated with the physical disfigurement, these characteristic changes can potentially identify the HIV status of the individual to others. This loss of medical anonymity amongst an oftenstigmatised population may contribute to an increased incidence of psychological illness. Patients often adopt coping measures similar to other patients with facial disfigurement such as disguising affected areas with hairstyles or clothes as well as avoiding social interaction. Importantly, some patients choose to stop their antiretroviral treatment.6 Current strategies for the management of facial lipoatrophy must include altering HAART regimes associated with lipodystrophy. Treatment options range from local flaps and autologous tissue-transfer techniques to correcting the deficit using biological or synthetic fillers which are injected locally.7 Whilst there has been a variable degree of success with different forms of fillers in terms of complications and permanence of change, polylactic acid (PLA) provides a promising medium-term treatment. A good cosmetic result for patients has the aim of normalising appearance, reducing fears of stigma and reducing psychological morbidity. The synthetic biodegradable implant PLA (Sculptra) is made of small (approx. 50-mm diameter) particles. The PLA has a long history of use in humans in the form of sutures, orthopaedic implants and more recently in sustainedrelease drug preparations. It is immunologically well tolerated and hydrolyses to lactic acid which is metabolised and ultimately excreted as CO2. These implants are injected as a suspension via a small-bore needle into the subdermal tissues. The particles act as a stimulus for fibroblast recruitment and as scaffold for the deposition of neo-collagen by these fibroblasts.8 The resulting filling effect is, therefore, independent of the lipodystrophic process.
J. Ong et al. Facial wasting in HIV-associated lipoatrophy often has an insidious onset and can go unnoticed by both patient and physician until a relatively advanced stage. The contours of the face are extremely complex and variable between individuals. Monitoring subtle volume changes over long time periods in an ageing population is particularly difficult. Facial laser surface scanning is repeatable and accurate9 and has been reported to have a higher sensitivity in measuring the subtle changes of HIV-associated facial lipoatrophy than clinical examination10 or facial anthropometry.11
Patients and methods Design This was a prospective study to objectively measure the efficacy of PLA in correcting the physical features of HIVassociated facial lipoatrophy and the effects of an aesthetic correction on psychological morbidity associated with facial disfigurement.
Sample One hundred consecutive patients with HIV-associated facial lipoatrophy were recruited from the Infectious Diseases Department at the Royal Free Hospital. All the patients were over the age of 18 years and had lipoatrophy grade confirmed by clinical examination by two doctors within the department (one doctor was usually the referring infectious diseases physician, the other being the plastic surgeon performing the treatment). Complete data are available on 65 patients. All patients were male, Caucasian, had no previous history of facial implants and had only had a single course of treatment. These treatments consisted of multiple small-volume (approximately 100 ml) injections of a suspension of PLA into the subcutaneous tissues of each cheek. The volume of suspension injected was adjusted according to severity of the lipoatrophy and to compensate for any asymmetry. Each vial was reconstituted to a volume of 5 ml, with 1.5 ml of lidocaine (2%) and 3.5 ml of water for injection. Two vials of PLA (Sculptra) were injected into each patient at each session. A firm facial massage was performed over the treatment area immediately after treatment. All patients were advised to massage their cheeks at regular intervals for 5 days. The patients underwent a total of four or five treatments at 4-week intervals. At each appointment, patients were assessed for any complications of treatment (e.g., bruising, infection and asymmetry). All the patients were also clinically reviewed 1 month following the last treatment and to ensure that all assessment measures had been completed and that followup had been confirmed. Ethical approval was granted by the local research ethics committee.
Measurement tools The patients were examined clinically using the regional scoring system used in the HIV Outpatient Study12 (HOPS)
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy and photographed. In this system, a score of 0 (none) represents no clinical facial lipoatrophy; 1, (mild) lipoatrophy detectable on close examination; 2, (moderate) lipoatrophy easily visibly to patient and doctor and 3, (severe) facial lipoatrophy obvious to a casual observer. These were recorded at 6-monthly intervals (pre-treatment, 6, 12, 18 and 24 months). Assessment of complex surfaces, such as faces, is difficult to perform accurately and objectively. We used 3D surface laser scanners as they are sensitive, quick to use and are comparable over time. By using the patient’s face as the baseline, we could provide an individualised internal control and thereby have a process which would allow an accurate detection of subtle changes in face surface. The 3D laser scans were performed on the same machine (University College London (UCL) optical face scanner) by the Maxillofacial Department of University College London. The patients were scanned pre-treatment and at 6-monthly intervals for 2 years during the follow-up period. These scans were then stored for comparison and analysis using Prima Facie software. The images were matched using fixed non-changing, untreated areas (e.g., forehead, ears, nose and chin). All scans were compared to the pre-treatment scan, and the volume changes were measured. Psychological factors associated with altered appearance were measured using the Hospital Anxiety and Depression Scale (HADS), the Rosenberg Self-Esteem Scale (RSE) and the Derriford Appearance Scale (DAS59). The HADS scale13 is a 14-item screening tool for anxiety and depression, standardised on both general and hospital populations. The HADS uses cut-off points as follows: normal (0e7), mild (8e10), moderate (11e15) and severe clinical range (16e21). The RSE14 is a short (10-item) scale, where low scores indicate high self-esteem. Normative data in the original test sample is skewed towards low self-esteem, but the instrument is sensitive to change and has been widely reported in the field of clinical psychology. The DAS5915 is a 59-item factorial scale measuring appearance-related distress, social anxiety and avoidance, standardised on both general and hospital populations. The DAS59 comprises subscales examining general selfconsciousness of appearance (GSC), social self-consciousness (SSC), self-consciousness of body and sexual selfconsciousness (SBSC), negative self-consciousness (NSC), self-consciousness of facial appearance (FSC), physical distress and dysfunction (PDD) as well as a whole scale (WS) score. Higher scoring on each of these subscales is associated with increasing psychosocial morbidity. The psychological measures of well-being (the Derriford subscales, subscales of HADS and self-esteem) were recorded at pretreatment, 6, 12, 18 and 24 months for each participant. All of the psychometric scales used in this study have good psychometric and scale properties. The multiple measures repeatedly measured on each participant may be analysed as a one-way, doubly multivariate, repeatedmeasures analysis of variance (ANOVA),16 with follow-up one-way, repeated-measures ANOVA for each measure. Application of Mauchly’s test of sphericity indicates that there is no basis for rejecting the sphericity hypothesis for the data collected and indicates that the analysis is valid. A pair-wise post hoc application of the paired-samples t-test
1629
on each measure, with Bonferroni corrected levels of observed significance, was used to locate precise differences in each measure over the five time points. The changes examined did not suggest gross deviations from normality, and, given the sample sizes in this study, the use of the paired-samples t-test is defendable as having robust validity and efficiency.17 The degree of the strength of the relationship between laser scores and each measure of well-being can be quantified using Spearman’s correlation coefficient. The laser scores at each 6-month interval represent a change from the baseline, and, for this reason, the strength of the relationship between laser scores and the corresponding change in each measure of well-being is also quantified using Spearman’s correlation coefficient.
Results Demographics All patients in this study were male and Caucasian. The mean age was 45 years, with a range of 30e72 years. The duration of infection with HIV was between 2 and 23 years (mean: 11.1 years). Of these patients, 46% had a clinical stage of acquired immunodeficiency syndrome (AIDS), 44% had symptomatic HIV infection and 10% had asymptomatic infection. The following analyses are based on a sample of n Z 65 patients who have provided data at each time interval. A missing-value analysis has been performed comparing baseline psychological measures between the group of patients with complete data (n Z 65) and those with incomplete data (n Z 35). These analyses indicate that there are no systematic differences between the two groups.
Clinical examination Mean values and standard deviations of clinical scores are included in Table 1. The distribution of severity of clinical lipoatrophy is illustrated in Figure 1, showing that clinical grades were either moderate or severe prior to treatment (mean: 2.82). These grades dropped to 0.66 (noneemild) after treatment (6 months) and progressively increased again to 1.26 at the 24-month review.
Clinical photography All patients had antero-posterior (AP), left- and rightlateral views of the face performed at 6-monthly intervals. These photographs were used as objective records of the scores given by clinical examination. Figure 2 shows preand post-treatment photographs of the same patient illustrated in the 3D laser scans.
3D laser scans Laser scans were performed before each treatment and before each subsequent 6-monthly review. Figure 3 shows computer-generated images of the laser scans. When the baseline scans (Figure 3, Left) are compared to posttreatment images (Figure 3, Right), a composite image is generated (Figure 3, Centre). The composite image shows
1630
J. Ong et al.
Table 1
Mean values and standard deviations of all measures at each time point
Measures of self-consciousness of appearance
Derriford Appearance Scale
Hospital Anxiety and Depression Scale Rosenberg Self Esteem Scale Change in volume over treated area (c.c.) Clinical Grade
Whole Score General self-consciousness Social self-consciousness Body/sexual self-consciousness Negative self-concept Facial self-consciousness Physical distress and dysfunction Anxiety Depression
Time (in months) Baseline
6
122.42 29.20 35.63 17.83
94.23 19.66 24.37 12.18
(35.10) (12.60) (13.93) (9.25)
15.23 (4.82) 5.09 (3.97) 2.48 (1.26)
16.34 (4.82) 4.12 (2.97) 2.49 (1.48)
10.25 (4.34) 7.51 (4.46)
7.23 (3.94) 4.00 (3.54)
8.68 (4.32) 5.35 (3.75)
9.02 (4.52) 5.05 (3.80)
10.26 (4.66) 5.28 (3.53)
22.05 (6.54)
18.45 (5.26)
19.97 (4.68)
20.55 (4.11)
21.94 (5.59)
2.81 (0.92)
3.00 (1.00)
2.99 (1.08)
2.68 (1.06)
0.66 (0.54)
0.87 (0.64)
1.06 (0.46)
1.26 (0.56)
Severe
70 2
60
10
9 20
Frequency
50 40
38 56
39 51
30
41 20 25
10
16
9 0 Baseline
6 Months
12 Months
(25.99) 109.68 (35.90) 113.46 (32.52) (8.76) 22.00 (8.68) 26.86 (9.65) (9.25) 29.14 (11.43) 29.12 (8.74) (7.05) 13.86 (7.89) 16.75 (8.83)
15.06 (4.08) 3.71 (2.67) 2.54 (1.43)
the differences between the surfaces of the baseline and the subsequent scan as a colour change. The scans from each time point were compared to the baseline (pre-treatment) scan to calculate volume changes over the treated area. The mean volume change from baseline to each time point is shown in Figure 4. The onesample t-test (scores at the baseline are all equal to 0) comparing all time points to the baseline shows that, after treatment the mean cheek volumes (per cheek) were all significantly different from 0 (mean difference Z 2.81 cc, p < 0.005). Differences in measurements of cheek volume compared to the baseline measurements and those between 6, 12, 18 and 24 months were assessed using the one-way repeated-measures ANOVA. A pair-wise application of the paired-samples t-test with Bonferroni corrected levels of significance confirmed that there were no significant differences in mean cheek
Moderate
98.66 20.31 26.37 14.17
24
13.52 (4.76) 3.22 (2.69) 2.69 (1.61)
2.82 (0.38)
Mild
(28.71) (7.48) (8.58) (7.95)
18
16.71 (5.05) 6.85 (7.79) 3.65 (2.17)
———————————
None
12
5
4
18 Months
24 Months
Time (Months)
Figure 1 The frequency distributions of clinical lipoatrophy scores over time.
projections between 6-, 12- and 18-month time points. However, a comparison of mean results between 18- and 24-month periods show a small, but statistically significant, decrease in mean cheek projection (p < 0.001).
Psychological measures Table 1 summarises the sample means (and standard deviations) for each measure for the five time points. Application of a one-way, doubly multivariate, repeated-measures ANOVA with the subscales of the DAS59 and measures of anxiety (HADS), depression (HADS) and self-esteem (RSE) as the dependent variables indicates that there are multivariate, statistically significant changes over the 24-month period (Wilks’ lambda Z 0.149, multivariate F(36, 29) Z 3.73, p < 0.001). Application of separate, one-way, repeated-measures ANOVA’s for each measure of well-being indicate that there is a statistically significant change in each subscale of DAS59, and in anxiety, depression and self-esteem over the five time points (p < 0.001 in all cases). A post hoc pair-wise application of the paired-samples t-test with Bonferroni corrected levels of significance for two-sided tests has been carried out for each measure. The Bonferroni adjusted p-values for comparisons at 6, 12, 18 and 24 months when compared with the baseline measures are summarised in Table 2. All measures of wellbeing show a statistically significant improvement after treatment at 6-month follow-up when compared with the baseline, except for physical distress (p Z 0.131). At the 24-month review, all measures of well-being, except for social self-consciousness (p Z 0.002), physical distress (p Z 0.004) and depression (p Z 0.028), returned to the baseline (pre-treatment) values. The pattern of statistical significance is particular to the measure under investigation (Table 2).
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy
Figure 2
1631
Clinical photographs before (left) and after (right) treatment.
Correlations between measures Laser scores (facial-volume changes) represent a change from the baseline scores. Accordingly, changes from the baseline in the measures of well-being were correlated with laser scores. At 6 months, only the change in facial self-consciousness was significantly correlated with the laser score (r Z 0.246, n Z 65 and p Z 0.048, two-sided). At 12 months, none of the changes in well-being was significantly correlated with the laser scores. At 18 months, the change in scores for negative self-consciousness and the laser scores did show a statistically significant correlation (r Z 0.313, n Z 65 and p Z 0.011, two-sided) but all other correlations were not significant. At 24 months too, none of the changes in well-being was significantly correlated with laser scores. Between the baseline and the 6-month time point, all participants showed an improvement in clinical scores coinciding with a change in the laser scores. Out of the 65 participants, 52 (i.e., 80% of the participants) showed an improvement on six or more measures of well-being. At 6 months, 25 participants had a clinical score of 0,
representing no clinical facial lipoatrophy. However, a multivariate ANOVA comparing those without clinical facial lipoatrophy with those showing some signs of facial lipoatrophy indicates that there is no significant difference on the measures of well-being between the two groups (Wilks’ Lambda Z 0.903, Multivariate F(9, 55) Z 0.583, p Z 0.821). For 27 patients, there were no changes in clinical grading between 6 and 24 months, whereas 35 patients showed a worsening in clinical grading. A multivariate analysis of the measures of well-being showed that the changes in measures of well-being between 6 and 24 months were not significantly different between those who did not have a change in clinical grading and those whose clinical scoring had increased (Wilks’ Lambda Z 0.778, Multivariate F(9, 52) Z 1.652, p Z 0.125). At 24 months, 20 patients had a clinical score of 2 (i.e., (moderate) lipoatrophy easily visible to patient), and the remaining 45 patients either had no clinically apparent facial lipoatrophy or had mild facial lipoatrophy that was only detectable on close examination. A multivariate ANOVA indicates that those cases showing moderate lipoatrophy that is visible to the patient at 24 months do not differ on the measures of
Figure 3 Laser-scan images showing the right cheek before (left) and after (right) treatment. The composite image (centre) shows the difference in scans represented by colour. Scale in mm is depicted at the bottom of the image. Note positive projection of skin surface of cheek (up to 8 mm) and negative change in beard area from shaving.
1632
J. Ong et al.
Figure 4 Ninety-five percent confidence intervals for change in cheek volume (cc) measured using laser surface scanning.
well-being from those cases with mild or no facial lipoatrophy (Wilks’ Lambda Z 0.902, Multivariate F(9, 55) Z 0.662, p Z 0.739).
Complications At each appointment, patients were asked specifically about complications from their previous treatment. Bruising was the most common complication, being reported after 7.5% of treatments. The only other reported complication was subcutaneous lumps. These occurred in 3.3% of patients, and all the lumps had resolved at the 1-year follow-up. All complications were treated conservatively.
Discussion Treating facial wasting with PLA implants is an effective way to normalise the disfigurement associated with HIVassociated facial lipoatrophy. In this study, patients had moderate-to-severe facial lipoatrophy, and treatment
reduced the clinical grade to a none-to-mild grade after four or five treatments. Previous studies have shown that psychosocial distress is related to physical deformity and facial disfigurement,18,19 although there is no evidence for a direct correlation between severity of disfiguring conditions and associated psychological distress. Facial lipoatrophy is not only disfiguring, but can also identify the HIV status of the individual in knowledgeable communities. The involuntary disclosure of HIV status can impact adversely on the well-being as a result of stigmatisation, discrimination and disruption of personal relationships.19,20 Treatment of HIV-associated facial lipoatrophy provides a naturally occurring AeBeA21 design with repeated measures taken before treatment (A), during treatment (B) and when conditions return to baseline (A). The treatment with PLA is successful in reducing the disfigurement. In addition, as the implant degrades (PLA is a synthetic polymer which hydrolyses and degrades by hydrolysis over time), we can examine the psychological impact of partially re-acquiring the deformity prior to starting a second course of treatment. Furthermore, facial wasting has little, if any, effect on normal physiology and any influence on social functioning is a reflection of the aesthetic deformity and its associated stigma. We found that mean 12-month and 18month cheek volumes are higher than the mean posttreatment (6-month) scans, thus showing continuing volume increase even after the end of treatment. This is illustrated in Figure 5, where improvements in cheek profile continue to progress after the end of the treatment process. Furthermore, although laser scans detected a significant drop in cheek volume from 18 to 24 months, the cheek volumes at 24 months were similar to the volumes immediately after treatment and significantly higher than baseline (pre-treatment) volume measurements. This can be explained by the relatively slow process of new collagenous tissue generation in response to the PLA implants, rather than direct volume increase due to the implants alone. (Patients generally only report subjective improvements after the third or fourth treatment.) Comparisons of the time points show that even at 24 months (although there is some volume loss from the peak at 12 months), there is still no statistically significant difference in mean volumes from those measured in the 6-
Table 2 Bonferroni corrected p-values from a pair-wise application of the paired samples t-test for each psychological measure between each time point and baseline Measure
Derriford Appearance Scale
HADS Rosenberg Self Esteem Scale
Time
Whole score General self-consciousness Social self-consciousness Body/sexual self-consciousness Negative self-concept Facial self-consciousness Physical distress and dysfunction Anxiety Depression
6
12
18
24
<0.001 <0.001 <0.001 0.002 0.001 0.013 0.131 <0.001 <0.001 <0.001
<0.001 <0.001 <0.001 0.167 0.538 0.033 0.010 0.576 0.050 0.462
0.252 0.004 0.022 0.094 0.925 1.000 0.002 1.000 0.015 1.000
1.000 1.000 0.002 1.000 1.000 0.123 0.004 1.000 0.028 1.000
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy
1633
Figure 5 Sequential clinical photographs illustrating facial changes; pre-treatment (left), post-treatment (centre) and at 12 months post treatment (right).
1634 month scans. This contrasts with a 0.6-point increase in clinical grades by the examining doctors between these two review points. Since it is unlikely that clinical judgement is more sensitive than the laser scanner in detecting a return in facial wasting, these data suggest a clinical bias towards placing a patient who may be of borderline grade in a classification of a worse clinical grade, particularly if the patient himself or herself perceives a change. Although the pattern of change was broadly similar between the different physical measures (laser and clinical), there were interesting differences in the psychological scales. The psychological questionnaires measuring appearance-related anxiety, anxiety, depression and selfesteem all showed benefits of treatment of facial lipoatrophy with PLA implants. Most of the scales had a similar pattern of change, namely a decrease in psychological morbidity after treatment, with a decrease in this improvement over time, but there were differences in the behaviour of individual scale scores during the review phase. The GSC, SSC, FSC and PDD had the most durable benefit, with mean values not returning to baseline levels at the 24month review point. Similarly, the improvements in the depression subscale of HADS also lasted up to 24 months. However, anxiety (HADS) scores fell to subclinical levels at 6 months, but reverted back to the baseline level at 24 months, moving back into the moderate clinical range. The SBSC and NSC returned to baseline levels at 24 months after treatment. The FSC levels dropped after treatment just as the levels in other subgroups of DAS did after treatment, but increased to levels reflecting higher social anxiety than the 24-month scores between the 12and 18-month reviews. At this point, there are no objective physical changes measured by laser scan measures, and mean clinical grades are still significantly lower than baseline grades. These scores, together with the elevation of the anxiety subscale of the HADS, support an explanation in terms of anticipatory anxiety. The patients are aware that PLA implants are not permanent, and the anticipation of a return in symptoms impacts on psychological wellbeing ahead of physical change. Anxiety is also likely to increase vigilance and scrutiny of the face, with patients who worry about the return of the disfigurement and its impact on their lifestyle, perceiving the onset of deterioration prior to its objective recurrence. This explanation is consistent with the observed bias in clinical grading for borderline cases. Analysis of laser-scan changes and psychological scores before and after treatment showed no correlation between the two variables in any of the psychological scales. This lack of correlation between the degree of physical deformity and level of psychological distress is consistent with the reports in literature and suggests that psychosocial adjustment is not directly related to the degree of deformity, but is modulated by individual factors such as the value placed on appearance, the subjective rating of visibility to others and the perceived likelihood of change. This study showed that treatment with PLA has a low incidence of complications. Bruising occurred after 7.5% of treatments. This was unpredictable and all patients had normal haematological parameters. All bruises settled spontaneously within 1e2 weeks. The most serious complication was the formation of temporary lumps. When
J. Ong et al. present, these were usually palpable at the post-treatment review and at the 6-month review. All had resolved by the time patients were reviewed 1 year after treatment. PLA implants improve the physical changes of HIV-associated facial lipoatrophy. They are easy to administer and produce a gradual improvement in facial profile over time. The absence of a sudden change also provides patients the opportunity to disguise the fact that they are having facial treatment. The physical changes shown by 3D surface laser are significant and last up to 24 months following treatment. However, the anticipation of a return in facial wasting at the end of treatment leads to a fall in psychological well-being when treatment ends despite negligible physical change measured on the cheeks. The clinical grading is more sensitive than laser scanning in predicting this change in psychological functioning. This is because it is driven by the objective assessment by the clinician as well as the subjective opinion of the patient and, hence, is more likely to be sensitive to any suggestion of change. Because of the impact of anticipated change, it is important to provide an explanation of the mechanism of collagen infiltration and reassurance that physical change will not be contiguous with treatment cessation. Re-assurance that treatment can be repeated to maintain appearance within the normal range will also be beneficial.
Acknowledgements The authors are very grateful for the kind help of all the patients and staff of the Plastic Surgery and Infectious Diseases Departments at the Royal Free Hospital who were involved with the clinical aspect of this study. Special thanks are also due to the medical imaging department at University College Hospital for their expertise in the use of the laser scanning equipment and help with the analysis of the images.
References 1. Palella Jr FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998;338:853e60. 2. Carr A. HIV lipodystrophy: risk factors, pathogenesis, diagnosis and management. Aids 2003;17:S141e8. 3. Carr A, Samaras K, Chisholm DJ, et al. Pathogenesis of HIV-1-protease inhibitor-associated peripheral lipodystrophy, hyperlipidaemia, and insulin resistance. The Lancet 1998;351: 1881e3. 4. Lichtenstein KA, Delaney KM, Armon C, et al. Incidence of and risk factors for lipoatrophy (abnormal fat loss) in ambulatory HIV-1-infected patients. J Acquir Immune Defic Syndr 2003;32: 48e56. 5. Barreiro P, Garcia-Benayas T, Soriano V, et al. Simplification of antiretroviral treatmentehow to sustain success, reduce toxicity and ensure adherence avoiding PI use 1. AIDS Rev 2002; 4:233e41. 6. Spire B, Duran S, Souville M, et al. Adherence to highly active antiretroviral therapies (HAART) in HIV-infected patients: from a predictive to a dynamic approach. Soc Sci Med 2002;54:1481e96. 7. Nelson L, Stewart KJ. Plastic surgical options for HIV-associated lipodystrophy. J Plast Reconstr Aesthet Surg 2008;61: 359e65.
Objective evidence for the use of polylactic acid implants in HIV-associated facial lipoatrophy 8. Gogolewski S, Jovanovic M, Perren SM, et al. Tissue response and in vivo degradation of selected polyhydroxyacids: polylactides (PLA), poly(3-hydroxybutyrate) (PHB), and poly(3hydroxybutyrate-co-3-hydroxyvalerate) (PHB/VA). J Biomed Mater Res 1993;27:1135e48. 9. Bush K, Antonyshyn O. Three-dimensional facial anthropometry using a laser surface scanner: validation of the technique. Plast Reconstr Surg 1996;98:226e35. 10. Benn P, Ruff C, Cartledge J, et al. Overcoming subjectivity in assessing facial lipoatrophy: is there a role for three-dimensional laser scans? HIV Med 2003;4:325e31. 11. Aung SC, Ngim RCK, Lee ST. Evaluation of the laser scanner as a surface measuring tool and its accuracy compared with direct facial anthropometric measurements. Br J Plast Surg 1995;48: 551e8. 12. Lichtenstein KA, Ward DJ, Moorman AC, et al. Clinical assessment of HIV-associated lipodystrophy in an ambulatory population. Aids 2001;15:1389e98. 13. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361e70. 14. Rosenberg M. Society and the adolescent self-image. Wesleyan University Press; 1989. 15. Harris DL, Carr AT. The Derriford Appearance Scale (DAS59): a new psychometric scale for the evaluation of patients with
16.
17. 18.
19.
20.
21.
1635
disfigurements and aesthetic problems of appearance. Br J Plast Surg 2001;54:216e22. Chaffin WW, Rhiel SG. The effect of skewness and kurtosis on the one-sample T test and the impact of knowledge of the population standard deviation. J Stat Comput Simulation 1993; 46:79e90. Kirk R. Procedures for the behavioural sciences. International Thompson Publishing; 1995. Stricker G, Clifford E, Cohen LK, et al. Psychosocial aspects of craniofacial disfigurement. A ‘‘State of the Art’’ assessment conducted by the Craniofacial Anomalies Program Branch, The National Institute of Dental Research. Am J Orthod 1979;76: 410e22. Harcourt DM, Rumsey NJ, Ambler NR, et al. The psychological effect of mastectomy with or without breast reconstruction: a prospective, multicenter study. Plast Reconstr Surg 2003; 111:1060e8. Chandra PS, Deepthivarma S, Jairam KR, et al. Relationship of psychological morbidity and quality of life to illness-related disclosure among HIV-infected persons. J Psychosom Res 2003; 54:199e203. Rosenthal R, Rosnow RL. Essentials of behavioural research: methods and data analysis. 3rd ed. McGraw Hill Higher Education; 2008.