As a service to members of the International Association of Physicians in AIDS Care (IAPAC), this issue of the IAPAC Monthly features a preview of "Treating Morphologic and Metabolic Complications in HIV-Infected Patients on Antiretroviral Therapy: A Consensus Statement of an Advisory Committee of the International Association of Physicians in AIDS Care," which will be published in the April/June 2005 issue of IAPAC's clinical journal, JIAPAC.
Objective: Clinicians are increasingly challenged by presentation of morphologic and metabolic complications in HIV-infected patients. These complications are associated with HIV infection and/or combination antiretroviral therapy. This Consensus Statement is intended to offer guidance to clinicians actively involved in HIV/AIDS care.
Participants: Seven clinicians with expertise in HIV medicine were invited by the International Association of Physicians in AIDS Care (IAPAC) to serve on an ad hoc Advisory Committee.
Consensus process: IAPAC convened the Advisory Committee to develop a draft Consensus Statement. Each clinician was tasked with drafting a specific section of the Consensus Statement corresponding with his or her expertise around a morphologic and/or metabolic complication. Scientific and clinical research, and other data in published literature and abstracts from scientific conferences were considered by strength of evidence. This document represents the consensus agreement of the Advisory Committee.
Conflict of interest disclosure: IAPAC sponsored and coordinated the development of this Consensus Statement with an unrestricted educational grant from Bristol-Myers Squibb. The opinions expressed in this Consensus Statement represent only those of the Advisory Committee.
Treatment of HIV-infected patients is frequently complicated by several morphologic and metabolic complications. These complications include:
Unfortunately, the mechanisms underlying these complications largely remain unknown. And, although various guidelines exist for treating HIV-infected patients with such complications, the recommendations offered are often not particularly user-friendly for the typical HIV care provider. In addition, these guidelines do not account for the compound effects of HIV infection or the effects that may be caused by altering and/or combining antiretroviral therapy (ART) with other therapeutic drugs. Consequently, an Advisory Committee of the International Association of Physicians in AIDS Care (IAPAC) was convened to propose practical recommendations for managing HIV-infected patients on ART who present with morphologic and metabolic complications, with the ultimate goal of optimizing treatment outcomes and improving patient quality of life.
Lipoatrophy is the most common morphologic abnormality associated with HIV and its therapy, and the highly distressing and potentially stigmatizing effects from this complication on patients' quality of life should not be underestimated.1 The typical presentation consists of facial, limb, and buttock wasting; prominence of the veins on the arms and legs; and greater definition of the individual muscles. Unlike HIV-related wasting seen before the availability of potent ART, lean body mass in patients with lipoatrophy often shows little or change.2
In a prospective cohort study comparing men with and without HIV, self-reported facial lipoatrophy of at least mild severity occurred in 42 percent of those with HIV on protease inhibitor (PI)-containing ART versus only 3 percent of seronegative controls. In contrast, fat accumulation occurred with equal frequency in both groups over time.3 The preliminary findings in the Fat Redistribution and Metabolic Change in HIV Infection (FRAM) cohort study4 strongly indicate that fat atrophy is the dominant component of the so-called "lipodystrophy syndrome," and that there is no relationship between fat atrophy and fat accumulation; hence, there is no redistribution of fat from subcutaneous stores to the abdominal visceral area as had previously been hypothesized.
Both patient- and treatment-related risk factors for lipoatrophy have been identified. The most important patient-related risk factor for lipoatrophy is the stage of HIV disease: patients who have a lower CD4 count at the time of starting treatment are at substantially greater risk.5 Other identified associations include duration of HIV infection and age, but lipoatrophy may still occur in a subset of patients in the absence of long-standing HIV infection or advanced HIV-related immunosuppression, even including those treated for acute HIV infection.
The best data regarding treatment-associated risk comes from prospective comparative clinical trials. A proposed hierarchy of treatment-associated risk for lipoatrophy is as follows: dideoxynucleosides -- stavudine (d4T), didanosine (ddI), and zalcitabine (ddC) -- carry the highest risk; zidovudine (ZDV) is intermediate; and tenofovir (TDF), abacavir (ABC), and lamivudine (3TC) should be associated with the lowest risk.6-8 The consistent observations that d4T-containing regimens are associated with a higher risk of developing lipoatrophy have led to a change in some treatment guidelines in which d4T is no longer recommended as a component of preferred first-line antiretroviral (ARV) regimens.9 This change was likely related not only to lipoatrophy, but also to other manifestations of mitochondrial toxicity. Although PIs have been more directly linked to insulin resistance and fat accumulation, the combination of nucleoside reverse transcriptase inhibitors (NRTIs) and PIs appears to act synergistically to accelerate fat loss.10 Other contributing factors are likely as well.11
The precise mechanism by which HIV therapy induces fat atrophy is still under investigation. The current leading hypothesis is that NRTI-induced mitochondrial toxicity induces fat cell apoptosis, with the risk greatest for those agents demonstrating the highest in vitro inhibition of the mitochondrial enzyme polymerase gamma. In addition, it has been demonstrated that HIV itself may lead to reduced cellular mDNA content compared with HIV-negative controls,12,13 rendering those with HIV more susceptible to the toxic effects of treatment. However, other studies suggest there is no relationship between mitochondrial toxicity and the development of lipoatrophy.12,14
Because the underlying pathogenetic mechanism(s) remain unclear, management strategies have met with limited success. Strategies consist of drug substitutions, insulin-sensitizing agents, and cosmetic surgery. Because these treatments are either relatively ineffective or costly, strategies to prevent the occurrence of lipoatrophy, such as selection of ARV regimens with low mitochondrial toxicity, are critically important.
Several clinical studies have shown that substitution of ABC for d4T leads to a gradual increase in limb fat that is often accompanied by a subjective improvement in facial appearance.15-17 Levels of mDNA content also improve in adipose cells, although peripheral blood mononuclear cells may or may not show improvement. It is not known if TDF would have the same effect, although this is presently under study. Substituting a nonnucleoside reverse transcriptase inhibitor (NNRTI) for the PI component of the regimen has had no consistent effect on morphologic changes.18
Three prospective clinical trials have evaluated rosiglitazone in HIV-related lipoatrophy; two showed no effect,19,20 and one demonstrated a modest but statistically significant improvement.21 Metformin is another insulin-sensitizing agent that has been studied in patients with HIV-related body habitus changes; however, treatment is associated with weight loss22 and hence should be avoided in cases of marked lipoatrophy.
Unlike the various pharmacologic interventions for lipoatrophy, which induce relatively small benefits, cosmetic surgery is often extremely successful and can dramatically improve appearance. Reported techniques include autologous fat transfer23 or, more commonly, injection of biologically inert substances such as polylactic acid.24,25 Patient satisfaction after polylactic acid injection is extremely high, and thus far the procedure appears safe. The major drawbacks to this treatment approach are the lack of long-term efficacy and safety data, the relatively high cost, and the lack of effect on lipoatrophy of the arms and legs. Patients referred for this procedure should be informed that most insurance policies and no state-funded programs cover the cost of this treatment.
Determining the incidence and prevalence of fat accumulation is difficult in the absence of specific diagnostic criteria, which do not exist. Based upon self-report with observer confirmation, about one third of treated, HIV-infected patients will show evidence of intra-abdominal (visceral) fat accumulation in the presence or absence of more generalized obesity.28 Between 5 percent and 10 percent of patients will demonstrate more generalized upper body obesity, including a prominent dorsocervical fat pad. Similar findings have been observed in both men and women, and both visceral and dorsocervical fat accumulation has been reported in children.29 There is some evidence in favor of racial and gender influences on prevalence. The interrelationships among fat accumulation and other manifestations of HIV-associated lipodystrophy are uncertain.
The pathogenetic mechanisms underlying the accumulation of fat in visceral or other fat depots are uncertain. Early observations showed evidence of fat accumulation in the presence of NRTI-containing therapy, although no potential mechanisms have been theorized to explain how NRTIs might promote fat accumulation. Some element of immune reconstitution, as opposed to ART, might be involved, but no study or hypothesis has specifically targeted a relationship to fat accumulation.
It is possible that visceral fat accumulation represents, in part, the development of the metabolic syndrome in those who are genetically predisposed. Various studies have estimated that between 15 percent and 25 percent of the population contains a genetic predisposition to developing the cluster of metabolic alterations referred to as the metabolic syndrome, and it is likely that a similar proportion of HIV-infected patients have these same genetic predispositions.
Aside from changes that would be obvious to any observer, there are neither accepted diagnostic criteria for visceral fat accumulation, nor are there generally accepted published normal values and measurements of these values. Hence, clinical estimation of visceral fat will be possible only by anthropometric means for the indefinite future, much as is the case for diagnosing the metabolic syndrome. It is likely that criteria adopted for diagnosis of the metabolic syndrome (e.g., waist size >102 cm in men) would be excessively conservative, because many HIV-infected patients have abdominal subcutaneous lipoatrophy such that a higher proportion of abdominal fat is visceral in HIV-infected patients. In one study, a composite criterion of waist circumference >88 cm and a waist/hip ratio >0.95 in men, plus waist circumference >75 cm and a waist/hip ratio >0.90 in women effectively detected patients with high visceral fat content, but also may have been excessively conservative.30 These criteria have not been subjected to independent validation, and further validation is required.
There are no known ways to avoid the development of visceral fat accumulation except perhaps with the prevention of weight gain. Switching ARV drugs has not been shown to be effective in reducing visceral fat content in the absence of disease progression. In patients with increased fat sufficient to produce symptoms, diet and weight loss may be tried. A weight loss of 8 percent to 10 percent may be associated with symptomatic relief and some reduction in fat, although excessive weight loss increases the risk of visible lipoatrophy. Pilot studies showed that resistance exercise may decrease trunk fat.31 Weight loss and decreased waist size were noted in a trial of metformin, which was used to increase insulin resistance.22 Recombinant growth hormone has been shown to decrease visceral fat content in several clinical studies, and the optimal dose and duration of therapy are currently being determined.32,33 Cosmetic surgery may be an option for the management of large dorsocervical fat pads, although the risk of recurrence exists. The US Food and Drug Administration (FDA) has approved no pharmacologic therapy for this complication.
Type 2 diabetes mellitus and insulin resistance, considered a precursor state to diabetes, are increased among HIV-infected patients.
Recent reports have identified an increased risk of impaired glucose tolerance, hyperinsulinemia, and Type 2 diabetes among HIV-infected patients compared to the general population.34-37 In most of the studies evaluating insulin resistance and diabetes in this population, traditional risk factors such as increased age, obesity, and being African-American or Hispanic remained significant. Furthermore, several studies have identified PI use and/or combination ART in general as conferring additional risk for diabetes development. Hepatitis C virus (HCV) infection is a known risk factor for Type 2 diabetes in the general population38 and has now been shown to increase the risk of diabetes among HIV/HCV-coinfected patients.39 Diabetes and impaired glucose tolerance are known independent risk factors associated with cardiovascular disease, and may be important factors in the long-term health of HIV-infected patients. In a large prospective study of more than 23,000 HIV-infected patients, a diagnosis of diabetes was associated with an increased rate of myocardial infarction.40
Several mechanisms may be responsible for the increased rates of insulin resistance and diabetes observed among HIV-infected patients. For example, PIs have been shown to induce insulin resistance in healthy, non-HIV-infected individuals and are implicated in dysregulation of glucose homeostasis in HIV-infected patients on ART.41 In addition, changes in body fat distribution, both increased central adiposity and peripheral fat atrophy, may contribute to insulin resistance in patients with HIV lipodystrophy.42,43
Given the increased risk of Type 2 diabetes identified in association with HIV infection and the use of ART, an annual fasting plasma glucose (FPG) test is recommended and should be performed in conjunction with annual fasting lipid profiles. A diagnosis of diabetes can be made by FPG >126 mg/dl or a glucose concentration >200 mg/dl following a two-hour oral glucose challenge. In both cases, a positive result should be confirmed with a repeat test on another day. There is no established clinical test for the diagnosis of insulin resistance; however, impaired fasting glucose (IFG) (fasting glucose >110 mg/dl and <126 mg/dl) and impaired glucose tolerance (IGT) (a two-hour glucose >140 and <200 mg/dl) are considered indicative of a pre-diabetic state. If IFG or IGT are present, or if additional risk factors such as HCV coinfection, obesity, and PI use are present, annual oral glucose tolerance testing is suggested. Fasting insulin levels may also be measured; however, there is considerable variability in insulin assays and no established norms among endocrinologists for clinical purposes.
The clinical management of Type 2 diabetes is the same as in the general population. Management should begin with recommendations and counseling on dietary and lifestyle modifications. Insulin-sensitizing agents should also be considered in the management of Type 2 diabetes in HIV-infected patients. A number of studies conducted in the context of lipodystrophy have demonstrated preliminary safety and efficacy in improving insulin sensitivity.19,20,22,34,44,45 Use of metformin can result in weight loss and therefore may be preferred for patients with relative central adiposity or obesity, whereas thiazolidinediones may provide improvement in fat atrophy.22,34,45 However, rosiglitazone has been associated with increases in low-density lipoprotein (LDL) cholesterol in both non-HIV-infected and HIV-infected patients and therefore should be used with caution in this population already at risk for hyperlipidemia.46
Although the benefits of ART are undisputed, there is growing concern regarding the association of increased cardiovascular risk and the dyslipidemias associated with HIV infection and ART.
Most studies clearly suggest that there is an increase in coronary artery disease (CAD) in HIV-infected patients compared with age-matched controls.47-53 Multiple observational cohort studies have been performed to determine if lipid elevations and metabolic perturbations induced by ART increase the rate of cardiovascular events. These studies, although not uniformly positive,54 have linked ART to a greater risk of myocardial infarction. Other studies have shown that ART is associated with a greater risk of dyslipidemia. However, some of the lipid increases after initiation of ART appear to be a return to baseline levels that were previously depressed due to HIV infection.55 It is clear that lipid levels in patients on ART frequently go beyond baseline levels, and high-density lipoprotein (HDL) levels remain lower than expected. In addition, short-term trials in uninfected subjects have demonstrated potent lipid-raising effects of several ARV drugs.56
Although dyslipidemia was originally attributed primarily to PIs, more recent studies have shown that specific NRTIs can also affect lipid levels. In particular, d4T has been found to produce greater elevations in cholesterol and triglyceride (TG) levels than TDF,57 and switching from d4T to TDF leads to improvements in lipid levels.58 In a large randomized trial in ART-naive patients, ABC was more likely to increase cholesterol and TG levels than ZDV.59 NNRTIs have also been associated with different lipid effects and generally result in elevations of total cholesterol without much effect on TG levels. In the 2NN study, nevirapine (NVP) was compared to efavirenz (EFV) and was found to be associated with larger increases in HDL cholesterol and larger decreases in the more favorable total cholesterol:HDL cholesterol ratio.60
Studies investigating the effects of HIV infection and/or ART on measures of vascular endothelial structure and function have been reported. For example, several studies have been conducted in HIV-infected patients and have shown greater carotid artery intima-media thickness (CIMT) in patients on ART compared with controls.61,62 One longitudinal study showed a more rapid progression in CIMT after one year in HIV-infected patients on predominantly PI-based ART compared to HIV-negative matched control patients.62 Cardiac computed tomography with coronary artery calcium (CAC) scoring has shown some negative results when HIV-infected patient cohorts have been studied;63 however, other studies have shown that patients on PIs had significantly higher CAC scores compared with untreated patients or patients on NNRTI-based regimens.64,65 In the aggregate, these surrogate marker studies are consistent with the large cohort outcome trials, and they validate the concern regarding potential premature development of atherosclerosis in HIV-infected patients on ART.
Lipid screening should be performed by means of an annual fasting lipid profile. Routine use of National Cholesterol Education Program (NCEP) guidelines should be considered to categorize patients into low, intermediate, and high risk of cardiovascular complications. (Table 1) For patients at high risk for CAD, exercise stress testing, preferably Thallium-type testing, should be considered in a similar manner as for non-HIV-infected, high-risk patients.
CIMT measurement is a sensitive indicator of atherosclerosis and adds predictive value to standard risk factor assessment tools.66,67 CIMT has also been used to track the progression or regression of atherosclerosis in response to therapeutic interventions.68 Performing CIMT is not feasible, however, for most providers; while CAC is more widely available. CAC determination is a sensitive indicator of the presence of coronary artery atherosclerosis,69 but whether it should be adopted as a screening test remains controversial. Patients referred for this procedure should be informed that most insurance policies and no state-funded programs cover the cost of this treatment.
The management of HIV-infected patients with dyslipidemia and/or coronary heart disease (CHD) should include monitoring and optimizing lipid levels through lifestyle changes, switching ARV drugs, and the utilization of specific lipid-lowering treatments. In HIV-infected patients, the use of lipid-lowering drugs may result in pharmacokinetic interactions with ARV drugs, thereby complicating clinical management of these patients. In the absence of randomized clinical trials, clinicians should aggressively treat atherogenic dyslipidemia by utilizing or switching to ARV drugs with the lowest potential to induce CHD. When such a treatment strategy is not possible or proves ineffective, clinicians should prescribe lipid-lowering therapy.
For HIV-infected patients at risk for cardiovascular disease, comprehensive reviews of guidelines have been published by several organizations,70-72 which align closely with recommendations for managing non-HIV-infected subjects at risk.73 The emphasis is on treatment of elevated low-density lipoprotein (LDL)-c modified by specific risk factors including cigarette smoking, blood pressure, HDL-c levels, family history of CHD, and older age.73 Treatment should start with therapeutic lifestyle changes including diet, increasing soluble fiber, reducing the intake of saturated fats and cholesterol, reducing weight, and increasing physical activity.
In patients for whom lifestyle changes fail, clinicians should first consider switching ART. Switching to a triple-NRTI-, NNRTI-, or atazanavir (ATV)-based ARV regimen (the latter being our preferred choice) has been shown to improve lipid profiles in patients who have developed dyslipidemia while on PI-based regimens.74,75 The likelihood of virologic failure has ranged from 10 percent to 15 percent in most published trials of ART switch studies.74 Unfortunately, at the present time there are no studies comparing the strategy of switching ART for dyslipidemia versus treatment with lipid-lowering agents. If deemed virologically safe, it is preferable to switch therapy, thereby avoiding increasing the burden of pills ingested daily, the increased costs of additional therapy, and the potential for adverse drug-drug interactions. Those patients who do require a PI-based regimen may consider switching to ATV (unboosted by ritonavir [RTV] in treatment-naive patients prior to current therapy, or RTV-boosted for the treatment-experienced) because of its lesser effects on lipid or glucose metabolism. Switching from a PI-based regimen to a triple-NRTI regimen is limited by higher rates of virologic failure, especially in those with prior non-suppressive ART or documented NRTI resistance.
Lipid-lowering therapy is recommended when the above measures fail to reach the desired targets. Treatment should be directed to the specific lipid disorder. (Table 2) The HMG CoA reductase inhibitors, or statins, are generally the first-choice therapy for elevated LDL-cholesterol.76 Statins typically result in a 30 percent to 50 percent reduction in LDL-cholesterol and a 20 percent to 30 percent reduction in TG levels. Due to recognized interactions of some of these agents with PIs, most authorities recommend initiation of pravastatin (20 mg daily) or atorvastatin at low doses (10 mg daily) with careful monitoring for adverse events.70,77 If no adverse events have occurred or lipid levels have not improved after six to eight weeks, the dose may be increased.71,78 Based upon numerous pharmacokinetic studies, simvastatin and lovastatin should not be used in HIV-infected patients taking PIs.72,79 The newer statin, rosuvastatin, is not primarily metabolized by CYP450 enzymes similar to pravastatin, but there are currently no data assessing its potential interactions with ARV drugs.80 Pravastatin levels decrease significantly (~50 percent) in the presence of most PIs, and doses may need to be increased to improve efficacy. A recent pharmacokinetic study demonstrated that simvastatin, atorvastatin, and pravastatin levels are decreased in the presence of EFV, suggesting that the lipid-lowering benefits of statins will be diminished in the presence of NNRTIs (EFV and NVP) that induce the metabolism of CYP3A4.81 Ezetimibe may result in a 10 percent to 20 percent reduction in LDL-c and has been shown to be synergistic when used with a statin.82
Fibrates (e.g., gemfibrozil) or fenofibrates are useful for the treatment of elevated TG levels. (Table 2) Treatment of hypertriglyceridemia should be based upon measurement of non-HDL-cholesterol levels (TC-HDL-c) that simultaneously take into account both atherogenic and cardioprotective lipid fractions. Fibrates generally induce reductions in TG by ~50 percent and increases in HDL-c by 5 percent to 10 percent (patients with very high TG levels may observe a rise in LDL-c when initiating fibrates). Fibrates are less likely to interact with PIs since they are not metabolized by the CYP450 system. Theoretically, they can be combined with statins to produce an additional TG-lowering effect. It appears that they may be appropriate in patients with hypertriglyceridemia and can reduce TG levels to normal (success observed in 64 percent of patients on PIs).70,74,83 Fenofibrate has been shown to lower TG by more than 50 percent and, when used in sequential combination with pravastatin, can help HIV-infected patients with combined hyperlipidemia reach NCEP Adult Treatment Panel (ATP) III goals.84
Niacin has not been shown to be safe in treating HIV-infected patients; pending the results of ongoing clinical studies in this population, it should not be used except in highly refractory cases. Studies show it does work, but at the expense of worsening insulin resistance.85 Fish oil (omega-3 fatty acids) may be effective in providing a 15 percent to 20 percent reduction in TG levels.86 An algorithm is provided to facilitate the management of dyslipidemia in HIV infection. (Figure 1)
Osteopenia/osteoporosis is a frequent problem among HIV-infected patients. Osteopenia carries a two-fold lifelong increased risk of fracture, whereas osteoporosis increases the risk four-fold.92 Other bone complications, such as avascular necrosis (AVN) of the hip, have also been reported.93
The frequency of osteopenia/osteoporosis varies between 28 percent in ART-naive patients94 to 40 percent to 50 percent in patients with more advanced HIV disease,95-97 making this one of the most frequent metabolic complications associated with HIV disease. HIV-infected patients frequently have other risk factors that predispose them to develop osteopenia.97 PIs were initially implicated in what was then considered to be a new side effect of therapy, but it is now clear that osteopenia and osteoporosis can occur in the absence of ART.98
The relative contribution of ART and HIV infection itself to the development of this complication, and its relationship with other metabolic complications such as lactic acidosis, lipodystrophy, and hyperlipidemia is still under study.99 The osteopenia/osteoporosis seen in HIV-infected patients on ART is associated with a state of high bone turnover with increased levels of bone formation and resorption.100
Because of the high frequency of osteopenia and osteoporosis in HIV-infected patients, serious consideration should be given to evaluating the bone mineral density (BMD) of all such individuals, especially if other risk factors are also present. BMD of the spine and hip is normally measured using dual energy x-ray absorptiometry (DEXA scans). The World Health Organization (WHO) defines osteopenia and osteoporosis by comparison to normalized measurements of BMD. The t-score is the number of standard deviations (SDs) that a specific measurement differs from the normal BMD of the population at 30 years of age. Osteoporosis is defined as having a t-score <-2.5 SDs. Osteopenia is defined as a t-score between -1 and -2.5 SDs.
For patients with established osteoporosis, a small study (100) has demonstrated the effectiveness of alendronate in combination with vitamin D and calcium, with increases in lumbar BMD equivalent to the ones seen in HIV-negative individuals (approximately 5 percent at 48 weeks). For patients with osteopenia, management should focus on lifestyle changes including discontinuation of tobacco use, increasing weight-bearing exercise, and an adequate intake of calcium and vitamin D together with ART. Other alternatives that have been used in the general population are not particularly useful in HIV-infected patients: estrogens can only be used in postmenopausal women, thiazides are only modestly effective and seem to work less well in males, parathyroid hormone might not be indicated in cases of high bone turnover, and calcitonin is very expensive and less effective than biphosphonates. Switching ART does not seem to have a significant impact on BMD.101
Asymptomatic elevation of blood lactate (hyperlactatemia), generally defined as a venous lactate concentration >2-5 mmol/L with a normal arterial pH, is common in HIV-infected patients on NRTI-containing ART. Lactic acidosis (LA), defined as a venous lactate concentration >2 mmol/L plus arterial pH <7.30 (or [HCO3] <20 mmol/L), although rare, is a very serious complication of ART. Moreover, LA with a venous lactate concentration >5 mmol/L is almost invariably symptomatic.
In general, asymptomatic hyperlactatemia has been found to occur in 8 percent to 21 percent of patients on NRTI-containing ART compared with 1 percent to 2 percent of untreated patients.102-104 Variability in assessing the incidence exists due to the technical difficulties in accurately measuring lactate levels and the lack of consistency in defining hyperlactatemia. However, when strict definitions/guidelines are used with uniform sampling techniques, the true prevalence has most recently been reported to be <3.5 percent.105 Symptomatic hyperlactatemia has been reported to occur in ~1.5 percent to 2.5 percent of patients with an incidence of four to 10 cases per 1,000 patient years of treatment, whereas LA is even rarer with an estimated incidence of 1.3 cases per 1,000 patient years.102,103,106
A small percentage of patients on NRTI-containing ART develop mitochondrial toxicity and hyperlactatemia or LA. Severe hyperlactatemia has been associated with all NRTIs, especially after long durations of exposure.102-104 It appears that d4T-containing ART is an independent risk factor for the development of LA;106 ddI-containing ART has also been implicated. Recently, it appears that concomitant use of ribavirin (RBV) and d4T/ddI places the patient at increased risk for LA development.102 Female patients, especially those who are pregnant, appear to be at greater risk of developing LA. Other risk factors include peripheral neuropathy and lipoatrophy, decreased bone mineral density, and obesity.102-104 Of note, a high proportion of patients with the HIV-associated neuromuscular weakness syndrome also have elevated lactate levels.107
Although still under investigation, the pathogenesis of elevated lactate concentrations appears to be related to the inhibition of mitochondrial DNA (mtDNA) polymerase by NRTIs. The resulting decrease in mtDNA leads to impaired synthesis of proteins, especially those in the oxidative/phosphorylation system. If >70 percent reduction in mtDNA occurs, the metabolism becomes anaerobic with the resultant generation of lactic acid. Recent in vitro studies have suggested that PIs as well as NRTIs may result in mitochondrial damage.108 In addition, other mechanisms may be operating; for example, ZDV appears to have direct toxicity on liver and muscle cells. In moderate to severe hyperlactatemia, the main target organ appears to be the liver.
Because the clinical features of hyperlactatemia are non-specific (e.g., fatigue, malaise, weight loss, nausea and vomiting, dyspnea, myalgias, and abdominal bloating), routine measurements of lactate levels are NOT recommended. Instead, a keen awareness of possible potentiating factors and vigilance are necessary. Lactate levels should be measured immediately in those suspected of suffering from LA, as multi-organ failure leading to dysrhythmias, respiratory failure, coma, and death may ensue. With LA, hepatic dysfunction predominates; hypovolemia and sepsis, however, are not seen. Other causes of LA such as pancreatitis, dehydration, and acute hepatic failure from other causes must, as well as sepsis itself, of course, be ruled out.
The mainstay of treatment is immediate cessation of NRTIs in all patients with confirmed lactate levels >10 mmol/L or symptomatic hyperlactatemia with levels >5 mmol/L. The remainder of the ARV regimen should also be stopped so as not to promote ARV drug resistance. Care is supportive with regular monitoring of lactate levels; however, the frequency of monitoring is not clear. Lactate levels may remain elevated for weeks to months after the resolution of symptoms. While efficacy has not been proven, adjuvant therapy with a number of vitamin coenzymes (thiamine, riboflavin, and vitamin C), electron acceptors (coenzyme Q), and L-carnitine used in other mitochondrial diseases may be helpful and carry little risk.102,104
Although it has been done successfully,108 reinstitution of NRTI-containing ART is not generally recommended for LA in accord with current treatment guidelines.70,71 If NRTIs are deemed essential components when re-starting ART, NRTIs with less potential for mitochondrial toxicity such as ABC, 3TC/emtricitabine (FTC), and/or TDF should be used, and routine monitoring of lactate levels may be indicated.
The treatment of HIV infection has been complicated by short-term toxicities and long-term complications. The latter were initially referred to as the components of the "lipodystrophy syndrome." These complications are more accurately referred to as the morphologic and metabolic complications of HIV disease and ART. Although these complications have been well described and characterized, the mechanisms underlying their occurrence remain elusive. Guidelines exist for helping the clinician manage these complications. However, current guidelines are not particularly user-friendly for the busy HIV care provider.
This Advisory Committee has endeavored to outline the major morphologic and metabolic complications, as well as to provide the essentials of their management for the front-line HIV/AIDS-treating clinician. Knowing that HIV medicine is always a rapidly changing field, the Advisory Committee recognizes that the approaches to dealing with these complications not only remain somewhat unclear, but also are subject to revision at any time. Nevertheless, it is the Advisory Committee's hope that the above information and recommendations will assist clinicians in better managing these complications.
Carl J. Fichtenbaum is with the University of Cincinnati; Colleen M. Hadigan is with the Harvard Medical School; Donald P. Kotler (Co-Chair) is with the Columbia University College of Physicians and Surgeons; Gerald Pierone Jr. is with the AIDS Research and Treatment Center of the Treasure Coast, Ft. Pierce, FL; Paul E. Sax is with the Brigham and Women's Hospital, Boston; Corklin R. Steinhart (Co-Chair) is with the Mercy Hospital, Miami; Pablo Tebas is with the University of Pennsylvania School of Medicine, Philadelphia.