icon-folder.gif   Conference Reports for NATAP  
  10th Conference on Retroviruses and Opportunistic Infections
Boston, Mass, Feb 10-14, 2003
Back grey_arrow_rt.gif
Fat and Bone: What D:A:D Says and Other Strangeness at CROI
David Alain Wohl, MD - The University of North Carolina
--Lipodystrophy and MI Risk (myocardial infarction)
--FRAM study: fat loss is hallmark of lipodystrophy syndrome; fat loss seen but often no fat gain in belly
--Buffalo Humps in HIV+ vs HIV-
--Lipoatrophy observed in NRTI-sparing regimen
--Cosmetic surgery for facial fat loss: New-Fill: encouraging results reported; Fat Transfer
--Bone Loss: in African-American women; Fosomax- encouraging preliminary study results; protease inhibitors
With an entire day of oral presentations dedicated to the issue of metabolic complications those of us who spend an inordinate amount of our time thinking and talking about fat, lipids, mitochondria, bone metabolism and such could only gloat while at the 10th Conference on Retroviruses and Opportunistic Infections that our day had come. However, while most of the data presented at the conference indicates that gains have been made in sharpening what has been a very fuzzy picture other results seemed to fly in the face of all we had previously thought, at times eliciting a collective 'huh?' from the attendees.
Lipodystrophy and MI Risk
The presentation at the conference that most helped in moving our understanding of the relationship between HIV therapy and metabolic problems was the reporting of the results of the large D:A:D Study, an analysis of data from 11 cohorts studies involving over 23,400 persons with HIV infection living in Europe, Australia and the US to determine the rates and predictors of myocardial infarction (heart attack) among these individuals (abstract 130). As opposed to the debacle of the World AIDS Conference in Barcelona last Summer when Barbaro and colleagues were forced to withdraw their suspicious results from a purported trial evaluating the effects of HAART regimens on cardiovascular disease events among HIV-infected subjects, the D:A:D Study had all the earmarks of a rigorously and well conducted study.
As described in greater detail by Judy Aberg in her report (Heart to HAART by Dr. Aberg- www.natap.org/2003/Retro/day19.htm ), this study demonstrated a linear relationship between duration of treatment with PI and/or NNRTI based combination HIV therapy and myocardial infarction (MI) and there are study limitations. The rate of MI increased 11 fold from 0.5/1000 patient years of follow-up to 6/1000 patient years of follow-up. The risk of having an MI increased by about 26% for each year of potent HIV therapy exposure. Other factors associated with MI, some of which carried a greater risk than HIV therapy, were identified and included age, male gender, previous cardiovascular disease history, smoking, diabetes and total cholesterol level. Triglyceride levels, nadir (lowest ever) CD4+ cell count and HIV infection duration were not associated with MI risk.
But here is where things get weird. Clinician defined lipodystrophy in this study was correlated with a reduced rather than a heightened risk of MI. Exactly what type of body shape changes the investigators were reporting was not clear (i.e. fat accumulation or wasting or both) but an actual negative association between changes in fat and cardiovascular disease at first glance is paradoxical as both fat accumulation and fat wasting have been associated with dyslipidemias. Further details and longer term data from the study may help explain this finding. Dr. Friis-M┐ller had no explanation for this baffling result but did end with an excellent point during her concise presentation of the data, that despite there being an increased risk of MI with HAART exposure, the risk of severe disease and death from these therapies pales in comparison with what untreated HIV infection can do.
Fat Wasting, Fat Accumulation and FRAM
Editorial note from Jules Levin: at the Intl Conference in Barcelona, Carl grunfeld's presentation on preliminary FRAM study results caught everyone by surprise when he reported finding that lipoatrophy was the main hallmark of the syndrome of body changes patients were experiencin; AND although some patients experienced fat gain in the belly along with fat loss in the periphery, fat gain in the belly was NOT experienced more on average in patients compared to a control group of patients. To the astonishment of many in the audience his preliminary study conclusion was that fat gain in the belly (vieceral fat deep in belly) was not a feature of lipodystrophy and perhaps should not be considered as part of a definition of lipodystrophy. The link to the review of his presentation at Barcelona is below. I do not think you can define lipodystrophy and there are individuals in the FRAM study and in real life who experience both fat loss in the periphery and fat gain in the viscera of the belly; and we cannot lose sight of that. As well, the FRAM study had limitations.
Another large clinical study of a metabolic complication with some surprising (read: controversial) results is the FRAM (Fat Redistribution and Metabolic Change in HIV Infection) Study. This is a cross-sectional study in which intensive metabolic evaluations and detailed chart reviews were conducted in several hundred HIV-infected men and women at 18 clinics across the US. The results of these evaluations were then compared with similar data obtained from a body shape substudy of the CARDIA Study, a prospective evaluation of risk factors for cardiovascular disease in the general population. To date, data reported from FRAM has focused only on men. There were three FRAM related posters each of which expanded on the preliminary data presented last summer at the World AIDS Conference.
The biggest take home message from FRAM, thus far, is that lipoatrophy (fat wasting) is the dominant body shape change experienced by persons with HIV infection. In a comparison of 412 HIV-infected men with 153 age matched HIV-uninfected controls, HIV+ participants reported more fat loss of their face, cheeks, arms, legs and buttocks than HIV- subjects - not a shocker for those who care for HIV-infected patients (abstract 732). But, in contrast to conventional wisdom, HIV-infected men reported less fat gain at the waist, neck, chest, back or abdomen than the controls. In fact, those who were HIV+ reported decreasing fat depots in the waist, abdomen, chest and back. On physical exam, peripheral fat was more often assessed as being reduced in HIV-infected men but they also were considered to have less central fat accumulation than HIV-negative controls (61% versus 82%). Self-report and physical examination also failed to detect an association between peripheral (limbs, face, buttocks) fat loss and central (trunk, breasts, neck) fat gain. Wasting of fat in the periphery, therefore, was associated with wasting, not gaining, of central fat.
DEXA scan and MRI data from the study were found on a separate poster but supported the lipoatrophy all over and not mixed picture message (abstract 733). HIV+ men with self-reported and clinician assessed peripheral fat loss had a lower body mass index, less limb fat and less subcutaneous upper trunk fat than HIV+ men without peripheral fat wasting. In comparison to HIV- controls those HIV+ subjects without peripheral fat loss did have lower body mass index, limb fat and subcutaneous lower trunk fat than controls even though they did not necessarily notice it. The decreased limb fat and lower trunk SAT in HIV+ men who do not have the clinical syndrome of lipoatropy compared to HIV- men suggests that clinical lipoatrophy underestimates the subcutaneous fat loss found in HIV infection. Visceral, or deep, abdominal fat was not significantly different between HIV+ men with and without peripheral fat wasting although those with peripheral fat wasting had significantly less visceral fat than controls. Again, a mixed picture of fat loss and accumulation (no matter what you may think, see or say) was not found.
Lastly, the FRAM group had reported that dorsocervical fat pad enlargement, ingloriously called 'buffalo humps' were no more common among HIV+ men than HIV- controls. Qualifying that preliminary result was a third poster (abstract 734) in which buffalo humps were assessed by physical exam and measured with non-stretch tape. Eight percent of the HIV+ men had a buffalo hump, according to the abstract, compared to a whopping 11.3% of the HIV-uninfected men. However, when HIV+ men had a hump it was much bigger than those found in HIV- men. The average size of a hump in HIV positives was 9.0 x 8.2 cm versus 5.5 x 5.4 cm in HIV negatives (a 2.5 fold difference).
Editorial note: This study compared 421 HIV+ men between the ages of 33 and 45 years from 18 centers to 151 similar age control men (HIV-) from the Coronary Artery Risk Development in Young Adults (CARDIA) study. MRI was performed to quantify subcutaneous adipose tissue (SAT) in lower (abdomen/back) and upper (chest/back) trunk and visceral adipose tissue (VAT). The study researchers reported that although buffalo hump is not more common in HIV+, it is associated with different fat distribution. They reported that having a buffalo hump may be a surrogate marker VAT (visceral adipose tissue: fat deep in the belly); that is having more fat in viscera in belly may be associated with being more likely to develop a buffalo hump. In HIV+ men, those with buffalo hump had higher BMI (body weight) than those without buffalo hump, whereas in HIV- men, BMI was similar whether they had a hump or not. In both HIV+ and HIV- men, those with buffalo hump had higher trunk SAT and VAT measured by MRI than those without BH. However, while HIV+ men with BH had less lower trunk SAT than HIV- men with BH, both HIV+ and HIV-men with BH had similar BMI, upper trunk SAT, and VAT.
There are important limitations of the FRAM Study that are fully described in the NATAP report of the preliminary data presented in Barcelona: www.natap.org/2002/barcelona/day33.htm
The investigators are planning to perform a longitudinal study of HIV-infected men and women which will address many of the shortcomings that are part and parcel of a cross sectional investigation. As was seen with the analysis of buffalo humps, there is a wealth of data collected in this study that when fully evaluated presents a picture that begins to more closely correspond with clinical experience.
Lipoatrophy Without NRTIs
Another interesting twist on the body shape change story emerged from a substudy of the Dupont 006 Study which compared AZT+3TC+efavirenz (EFV) versus AZT+3TC+indinavir (IDV) and EFV+IDV (abstract 737). In the substudy, reported by Karen Tashima, single slice CT scans of abdomen were performed on 373 subjects remaining in the study when the substudy was initiated. The mean time from start of study antiretrovirals to the first substudy CT scan was 738 days. A second CT scan was done a year after the first. Changes in fat distribution were compared between those taking nucleoside reverse transcriptase inhibitors (NRTIs) (AZT+3TC+ either EFV or IDV) versus NRTI-sparing regimens (EFV+IDV). Almost 90% of subjects demonstrated a decrease in subcutaneous abdominal fat between scans regardless of whether they were receiving NRTIs or not. Over half the subjects had decreases in visceral abdominal fat during the substudy, again regardless of the presence of NRTIs in their regimen. These data suggest that lipoatrophy can occur independent of NRTI exposure, buttressing arguments that HIV itself and/or immune reconstitution may play a significant role in fat changes. Importantly, this substudy was performed almost 2 years after the subjects had started HIV therapy and involved AZT rather than d4T, the NRTI that is most commonly, albeit hardly exclusively, linked to lipoatrophy.
New-fill and 'Chipmunk Cheeks'
That lipoatrophy reigns as the predominant body shape change seen in HIV-infected persons in the era pf HAART is clear. However, there have been few therapeutic options available to treat fat wasting despite the disfigurement many suffer. A few posters described the results of cosmetic procedures to restore the facial features of HIV-infected persons with fat wasting. One of the most exciting interventions is polylactic acid or New-fill. This compound is injected into the skin at certain angles and is described as leading to long term thickening of the skin sufficient to fill the hollows left by the loss of facial adipose tissue. In a follow-up report on a series of about 50 patients receiving New-fill, cosmetically satisfying results were seen almost years out although some patients did require periodic reinjection (abstract 719). This agent is not approved for use in the US but is in parts of Europe.
Surgical treatment of facial fat loss with fat transfer from other parts of the body was not always so pretty or possible according to a series of 25 patients undergoing such surgery (abstract 722). A significant proportion of subjects did not have sufficient fat depots to transplant and some developed lipohypertrophy (fat enlargement) during HIV treatment of their transplanted fat leading to another zoonotic moniker for a body shape alteration, "chipmunk cheeks". Overall, though, many subjects were satisfied with the surgery. Surgical correction of buffalo humps in 26 patients with ultrasound assisted liposuction was reported by a surgical group from California (abstract 721). Results were mixed. Fat accumulation involving the posterior neck responded better than anterior fat and there were a number of complications including infection and mandibular palsy. Recurrence also was a problem with 7 subjects having a significant return of fat accumulation. Despite these problems, most patients seemed glad they underwent the procedure - on a scale of 0 to 5 with 5 being best, the mean satisfaction score after several months was 3.8.
Bone Mineralization and PIs - It's not what you think
Decreases in bone mineral density have been found to occur at unexpectedly high rates among HIV-infected persons. Although once considered to be a complication of HIV therapy, other factors for accelerated bone loss, including traditional risk such as low body weight, corticosteroid exposure and smoking, as well as HIV itself have been implicated. In a study of 42 HIV+ women who were on average about 40 years of age, 55% African-American, 84% smokers, largely overweight and generally improvised followed for 2 years, whole body DEXA scanning revealed scant (0.5% per year) bone density loss overall (abstract 102). The average CD4+ cell count for the group was almost 500 cells/mm3 and the mean viral load around 3000 copies/mL. Independent risk factors for decreased bone density during the study were low albumin, a 5% or greater weight loss and an increase in CD4+ cell count. No association between bone loss and HIV therapy was detected.
Another study also aimed at identifying risk factors for bone loss in HIV+ women was presented and also found little difference in DEXA defined bone loss of the hip and spine among a cohort of 200 HIV-infected, predominantly African-American women compared to HIV negative controls (abstract 103). About 30% of the HIV positive women developed osteopenia or osteoporosis versus 24% of the HIV negative control subjects - a difference that was not statistically significant. Age, white race and low body mass index were all independently associated with risk of bone loss. However, protease inhibitor (PI) use was associated with a ╔ (you will never guess) an increase in bone density. Among those with less than a year of PI exposure 36% developed osteopenia or osteoporosis versus 25% of those with more than a year of PI use. Go figure.
Lastly, on the topic of bones, a pilot study looking at the use of alendronate (Fosomax) to improve bone mineralization among HIV-infected subjects with osteopenia/osteoporosis (T-score less than -1) was presented by Pablo Tebas. For more information on T-scores, Z-scores and HIV related bone problems go to: http://aactg.s-3.com/metabolic/default.htm
In this small open label trial (abstract 134) 31 subjects (89% male, 80% white, 29% tobacco addicts, 61% on a PI, 16% moderate alcohol use) were randomized to alendronate + 400 IU vitamin D + 1000 mg calcium versus the vitamin D and calcium alone. Over the course of 48 weeks, there was a 5.2% increase in bone density at the lumbar spine in the alendronate arm compared to a 1.3% increase in the vitamin D + calcium alone arm (p=0.005). However, there were no statistically significant differences in the change of bone mineralization of the hip, femur or in overall bone mineral density. Alendronate was well tolerated with no treatment limiting toxicity experienced. A larger placebo controlled study of alendronate with equal representation of women and men is soon to begin within the AIDS Clinical Trials Group (ACTG).
The unexpected findings from several studies of metabolic complications clearly demonstrate the one certainty we have come to appreciate regarding these problems - they are complicated. That multiple factors besides HIV therapy influence development of body shape changes, is increasingly evident. Despite this, or perhaps because of it, our understanding of the epidemiology and pathogenesis of a number of metabolic conditions associated with HIV and its treatment has grown. The curve balls that pop out every once and awhile keep us swinging away until we score with better strategies for preventing and treating these troublesome complications.