Answers are grounded in this paper and cite their sources.
Introduction In the developed world, the HIV epidemic has entered its third major phase. Initially patients infected with HIV routinely succumbed to complications of opportunistic infections such as Pneumocystis carinii pneumonia, infection with Mycobacterium avium complex, or cryptococcal meningitis. In the second phase of the epidemic, better understanding of the natural history of HIV infection and the need for, and availability of, effective primary prophylaxis against the common opportunistic pathogens altered the mortality in individuals infected with HIV. Morbidity and mortality in this phase was associated with complications such as infection with cytomegalovirus, periventricular multifocal leukoencephalopathy, opportunistic lymphomas and/or the HIV-associated wasting syndrome [1]. Currently, in the era of highly active antiretroviral therapy (HAART), in which it is possible to maximally suppress the replication of HIV in the peripheral blood in a large percentage of HIV-infected individuals for a select period of time, the clinical profile of the complications of infection with the virus is changing again. Early in this third phase, reports of metabolic and morphologic abnormalities have appeared with increasing frequency, even or particularly, in those patients who have often had the best immunological or virological response to antiviral therapy [2-13]. Whether these changes occur as a direct result of the antiretroviral therapy (that is an adverse drug effect) or whether these changes occur because the antiviral agents have allowed patients to survive to the point in time at which this syndrome would naturally develop is not clear. These metabolic complications have included glucose intolerance [which appears to be a complication of the use of protease inhibitors (PI)], abnormalities in lipid metabolism, as well as a syndrome involving a change in body shape with a redistribution of body fat, often called ‚peripheral lipodystrophy syndrome‚ which has been much more difficult to describe or understand. All of these metabolic complications are the subject of this review, which will attempt to clarify the current knowledge regarding the epidemiology and clinical presentation of these complications as well as to summarize the difficulties that exist in defining the fat redistribution syndrome. It will also address the directions that future clinical and basic investigations may take in attempting to understand these phenomena better, so that directed interventions may be conceived and implemented. These abnormalities, which, in this review, will be called ‚the fat redistribution syndrome‚, have been variably called ‚protease paunch‚, ‚Crix belly‚, ‚lipodystrophy‚, ‚peripheral lipodystrophy‚, ‚pseudo-Cushings disease‚, ‚pseudo-cachexia‚, the ‚truncal obesity syndrome‚, ‚buffalo hump‚, ‚symmetric lipomatosis‚ among others. As will become apparent, all of these names are misleading and/or incomplete. Although the clinical presentation of the syndrome is variable, the major clinical features that have been recognized are listed in Table 1. The term ‚fat redistribution‚ will be used to include the body morphological abnormalities as well as the metabolic disturbances, although it is not clear at present that these are all necessarily components of the same syndrome.Table 1: Recognized clinical features of the HIV-associated fat redistribution syndrome.These morphological abnormalities are not all uniformly present in all individuals who present with fat redistribution syndrome. The frequency of each of these clinical features is not clear, nor is the frequency with which any of these clinical features must or may be seen in association with any other of the clinical features. Therefore, at present, it is not clear whether there are one or more syndromes which are being recognized and described simultaneously. Specific documentation of each of these features is difficult in clinical practice. Whereas patient weight is measured routinely in visits to primary care providers, it is a very non-specific measure and will not necessarily predict or define the patient with this syndrome. Waist-to-hip ratios, breast size, or mid-arm and mid-thigh circumferences, which would be more pertinent to the identification of this syndrome, are rarely or never measured in HIV-infected individuals in the primary care setting. Even in research settings, measurements of buffalo humps and/or regional body fat distribution are difficult to perform accurately. An accurate assessment of measures of facial fat loss or of prominent veins is also quite difficult. Bioelectrical impedance analysis, which is the most routinely available measure of body composition will not detect the fat re-distribution that occurs with this syndrome. Patient awareness of an alteration in body and/or facial shape is often the initial indicator of the presence of this syndrome, and the patient often calls attention to it [3]. As awareness of this fat redistribution syndrome has been increasing among physicians, it has become increasingly clear that there is tremendous difficulty in providing accurate documentation of the clinical features of this syndrome and accurately identifying all of those patients who may have some of the clinical manifestations. Recognition of the syndrome may occur late, unless a patient is presenting with a prominent buffalo hump. Even the primary care providers who have taken nutritional and metabolic issues seriously in patients with HIV and have monitored body composition routinely for individual patients will not necessarily be able to accurately identify these patients early in the course. The laboratory findings that occur with the syndrome are equally problematic (Table 2). Individuals who present with this clinical syndrome are often evaluated for Cushing‚s disease [3,9,10,14,15]. The vast majority of them have been found to have normal cortisol levels, whether random levels or 24 h urinary cortisols are checked. Dexamethasone suppression tests have also been normal. Similarly, the vast majority of individuals with the syndrome have blood glucose levels that are within normal limits. In the cases in which glucose tolerance or insulin resistance is evaluated, these values may be abnormal, but gross measures of blood glucose are rarely out of the normal range. Serum hormone levels other than cortisol are also rarely abnormal: thyroid, testosterone and estrogen levels have been within normal limits when these values have been determined. However, the blood lipids are usually significantly abnormal. Serum triglycerides and cholesterol levels have been elevated in a large percentage of fat redistribution syndrome in whom these levels have been determined. In the usual patient with HIV, serum triglycerides may well be abnormally high, but blood cholesterol levels are usually in the low to normal range. The elevations in cholesterol, therefore, separate the patients with this metabolic derangement in the fat redistribution syndrome from the usual patient with HIV [16].Table 2: Recognized laboratory features of the HIV-associated fat redistribution syndrome.The result of all of these factors is that the epidemiology of the fat redistribution syndrome remains obscure. Developing an understanding of the etiological associations has been complicated by a lack of data on the epidemiology of the syndrome. Fat redistribution and its associated body shape changes, however, are great concerns for patients with HIV who are doing well on their antiretroviral regimens. The data that currently exist to assist us in understanding the presentation of the syndrome will be summarized here. Epidemiology Body shape changes There are few prospective studies that look critically at the prevalence of the various reported or documented body shape/composition changes and the laboratory changes in the fat redistribution syndrome. The majority of the reports are retropective descriptions from a particular HIV population or clinic, so that bias in the selection and description of patients is unavoidable. Early observations noted the occurrence of buffalo hump and subsequently descriptions of the remainder of the clinical features of the syndrome began to appear. In one of the initial reports of buffalo hump, four out of eight (50%) of the patients had not taken PI [14]. In retrospective studies looking at the prevalence of the fat redistribution syndrome, various investigators have arrived at a prevalence of between 0.1% and 64% of HIV-infected patients taking PI [2,4,6,17-24]. These initial observations also report the development of these morphological abnormalities over a wide range of time after the initiation of HAART. At the 12th International AIDS Conference, additional epidemiological data appeared. Studies reported a range in prevalence: 24% of 29 patients developed fat redistribution by 3 months after initiation of PI therapy; 21 out of 116 (18%) women treated with PI developed changes in body composition 2-22 months after beginning PI therapy [25] and of these 20 were taking indinavir, 90% had an increased waist circumference, 71% increased breast size, 43% exhibited peripheral wasting but only 19% developed buffalo humps. These are unique data in that they describe the syndrome in women; they are also unique in that they provide data on the frequency of the individual clinical components of the fat redistribution syndrome. In the literature and by anecdotal report, buffalo humps appear to be less common than the remainder of the body shape changes reported. One investigator suggests that buffalo humps do not develop but appear as a result of the wasting of neck and truncal fat around the dorso-cervical area [26]. To explore further the occurrence of the specific components of the syndrome, a published report identifies the appearance of facial wasting in seven out of 29 (24%) patients within 13-35 weeks of starting indinavir therapy [27,28]. Other studies suggest that as few as 1.8% (five out of 272) of PI-treated patients develop the body shape changes, when the presence of changes is determined purely on a subjective basis [29]. Several studies suggest that the accuracy of patient identification of the syndrome is quite good [3,17]. Additional studies that attempt to quantitate the body shape changes [by regional dual energy X-ray absorptiometry (DEXA)] suggest that 12-18% of PI-treated patients develop these body shape abnormalities [30]. The highest prevalence of the fat redistribution syndrome was noted in an Australian study which suggested that 64% of HIV-infected patients treated with PI therapy develop body shape changes [17]. Although all of these studies have suggested that the appearance of the fat redistribution syndrome is strongly associated with the use of PI, in many of the studies only patients treated with PI therapy were surveyed for the presence of fat redistribution. In one of the first reports of the appearance of buffalo humps there were equal numbers of patients taking PI and those that were PI naïve [14]. In the ustralian study that found the highest prevalence of the fat redistribution syndrome, one out of 29 (3.4%) patients in the PI-naïve group used as controls had symptoms consistent with fat redistribution [2,17]. It is possible that because of the focus of caregivers and patients on the occurrence of this syndrome in individuals taking PI, that the body shape changes are more subtle and/or underreported in patients who are not being treated with PI. Laboratory abnormalities Lipids and cholesterol There have also been a number of studies examining the epidemiology of laboratory abnormalities seen with the fat redistribution syndrome. A study in a group of 340 HIV-infected patients, of whom 95 were not being treated with antiretroviral agents, 87 were receiving antiretroviral agents but not PI and 158 were taking PI found that cholesterol levels were increased in both of the treated groups [31]. The group treated with PI had higher levels of cholesterol; 16 % had cholesterol levels of greater than 65mg/dl and 5.7% of these patients had cholesterol levels greater than 100mg/dl. In another study, cholesterol, triglyceride, uric acid, and glucose levels were examined in 131 patients treated with PI and 25 patients who were treated with non-nucleoside reverse transcriptase inhibitors (NNRTI) [32]. Cholesterol, including high density lipids, triglycerides and uric acid were increased in those patients taking PI as compared with those taking NNRTI. There was no difference between the two groups in glucose levels. This is one of the few studies that reports body shape data as well as lipid abnormalities. In this study, 18% of the patients taking PI also had body shape changes consistent with fat redistribution. Ritonavir appears to be the PI most strongly associated with these abnormalities in lipids [33,34]. The reported incidence of increased cholesterol or triglycerides ranges from <5% to >90% and probably reflects the selection bias of patients reviewed for these abnormalities. The true incidence remains to be determined [2,35]. Glucose intolerance Abnormalities in glucose metabolism have also been documented in patients using PI therapy [36-38]. Of 783 patients taking PI 84 (11%) had glucose levels >140mg/dl and 3.3% (26/783) had glucose levels >180mg/dl [40]. In another study 15% of patients (45/295) had glucose levels >120mg/dl, although only two patients in this study were considered to have frank diabetes [40]. Other studies have documented elevated blood glucose in 11.5% (25/216) and 2% (26/1300) of patients taking PI [41]. Intravenous glucose tolerance tests were performed in 67 patients taking PI, 13 PI-naïve patients, and 18 HIV-negative patients to determine peripheral insulin resistance [42]; 61% of the patients taking PI exhibited insulin resistance, those taking indinavir were more severely affected than patients taking other PI. Cardiovascular complications There is a well known association of truncal obesity with risk for cardiovascular morbidity [43]. The abnormalities in lipid and glucose metabolism as well as the disturbances in body fat raise the concern that the fat redistribution syndrome may be associated with increased atherosclerotic complications. A report of two myocardial infarctions in young males with HIV (aged <40 years) treated with PI has been published, and an additional two myocardial infarctions have occurred in three young males with documented severe coronary artery disease [44,45]. These patients had elevated triglycerides and cholesterol. These observations prompted investigators to look at the levels of cholesterol in 89 patients treated with PI. They found that the mean cholesterol level of 270mg/dl was significantly higher than that seen in HIV-infected patients who were not taking PI (161mg/dl). Pancreatitis Elevations in triglycerides associated with PI in this population have also been reported to cause serious pancreatitis, but this has been less well reported in the literature [46,47]. Measurements of body composition Usual measures of body composition such as bioelectrical impedance analysis have not been helpful in defining body composition changes in the fat redistribution syndrome because this analysis documents total body lean mass and from that, total body fat. Bioelectrical impedance analysis is not able to detect the regional distribution or re-distribution of body fat. DEXA performed in the routine fashion is also not able to detect the distribution of fat to various regions of the body. However, DEXA results can be recalculated to determine regional fat compartments. At this time, the use of DEXA for analysis of regional body fat remains experimental. Regional body composition can also be determined using imaging studies such as computer assisted tomography or magnetic resonance imaging (MRI), which can visualize and quantify body fat [48]. It should be noted that there is no validation of DEXA as a measure of regional body fat by comparison with total body MRI or computer assisted tomographic scan. In addition, results of DEXA may vary with the scanner used, and standardization of scanners is not performed routinely. Therefore, any data generated by using this technique must be reviewed with caution. By using regional DEXA, total body potassium (40K) for the determination of lean body mass and whole-body MRI, investigators have been able to quantify fat deposition in selected body regions [49]. The regional body composition in 26 HIV-infected patients (15 men and 11 women) was compared with that of 26 controls. Seven of the HIV-infected men and five of the HIV-infected women reported an increased waist circumference, and relative wasting of arms, legs and face. There was a trend toward a decrease in total body cell mass in the HIV-infected patients; DEXA revealed a decrease in total body fat in the HIV-infected men and decreased fat in the limbs of the HIV-infected men but not in the women. MRI revealed a decrease in total body fat in the HIV-infected men and a decrease in the subcutaneous fat in the HIV-infected men; again these changes were not seen in the HIV-infected women. Visceral adipose tissue was also increased significantly in the HIV-infected men but not in the women although there was a trend toward an increase in the small number of women studied. Those patients who had reported an increase in waist size showed no change in body cell mass or lean mass by DEXA when compared to those who did not report a change in waist However, in these patients, the men were noted to have increased total body fat and increased adipose of to subcutaneous fat and to total body fat were increased in both men and women who reported increased waist of body composition by measures between HIV-infected patients on PI HIV controls and HIV-negative controls have that the current HIV-infected patients had a greater mean body weight and increased body fat over the HIV controls. These measures were in both of the HIV-infected however, than in the HIV-negative controls. analysis of the HIV-infected groups were but significantly from the HIV-negative controls of fat redistribution syndrome with HIV The of is that the fat redistribution syndrome is more severe in patients who have better to antiretroviral therapy with in cell and of HIV One study has reported the association of increase in cell with fat and decreased [17]. from a number of suggest that there is no association between cell and occurrence of the fat redistribution syndrome. from showed that of the patients with the fat redistribution syndrome had a but had a of these patients were not taking PI history of the fat redistribution syndrome The natural history of this syndrome remains to be determined. However, data suggest that the syndrome and not treated in some of fat redistribution syndrome There is a of among caregivers and patients the to the fat redistribution syndrome. are in Table To there are no of any therapy and all reports of therapy are of for the fat redistribution syndrome has been suggested in of the buffalo hump of hump after has been reported. to attempt to decrease fat has not been because these are than of fat is the antiretroviral has been only in body shape to the In 21 patients with fat redistribution who were from indinavir to reported further of the fat reported no change in their body and reported in blood lipids were noted in the patients therapy was but there was no change in blood cholesterol levels with the change in of increased cholesterol with or has also only been in triglycerides and cholesterol [44,45]. that any agents used in the of this syndrome should be a that is not by the or In those patients who have PI blood lipid levels have significantly In another study in which PI were body shape changes did not resistance has been as a to decrease fat and increase peripheral but has not been in a to The results of have been documented in two women in whom had a there was an increase in peripheral and no change noted in or breast size The use of hormone has also been as a therapy for the fat redistribution syndrome. This has not been to although have to In two patients with fat including buffalo hump, truncal peripheral elevated triglycerides and weeks of hormone to a of the body shape changes in one patient and a of the body shape changes in the patients had in triglycerides and cholesterol with hormone glucose is There was no change in body weight in patient have been suggested as possible but no data on their or have been patients receiving testosterone therapy and patients taking testosterone for weight have developed the fat redistribution syndrome on At present, the best to patients is not to PI therapy at the time of the identification of the fat redistribution syndrome of patients to research on this syndrome is being will assist in the of good data and the understanding of the syndrome. At present, there are more the fat redistribution syndrome than there are data or A is A better understanding of the epidemiology of the syndrome is as well as an understanding of the of a patient for It is not clear at this point in time that this is a syndrome, it may be that there is a of metabolic syndromes which have and which will develop in patient Although the weight of the to suggests a association with the use of PI it may be that specific components of the syndrome are associated with specific or and that other components are It appears that the lipid abnormalities and the glucose intolerance will be found to be associated with the use of the association of body shape changes with the use of PI is less clear. There may be selected patients among those that take PI therapy that are more at risk for the fat redistribution syndrome. as and/or nutritional may also the development of the syndrome. It may become clear that features are for this syndrome and will as clinical to for the development of the syndrome. The natural history of the fat redistribution syndrome is not clear. individuals appear to have an increased risk for coronary artery This is an association that to be and to attempt to any further morbidity and The for the majority of individuals who body shape changes but do not appear to have coronary to be determined. The for individual patients who some but not all of the clinical syndrome are not The of this is The that the protease has with on the low density lipid and on acid suggests that there may be a metabolic basis for some of the changes seen have suggested that there may be in cortisol metabolism, by HIV as the of HIV is known to the on some or by some of the HIV a of the fat redistribution syndrome is and prospective epidemiological studies are able to the clinical and laboratory of the syndrome, it will be more to attempt to understand the It is that more than a will be found for all of the clinical and laboratory features of the syndrome. better understanding of will the development and of interventions that may assist in the of and for the individual infected with HIV.
No takes yet. Share an insight, caveat, or question.
Christine Wanke (1999) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: