Issue: November 2010
November 01, 2010
14 min read

Physicians face increasing metabolic, CV complications in the treatment of HIV patients

Issue: November 2010
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

HIV-related morbidity and mortality has substantially decreased since new HIV therapies emerged more than a decade ago. However, physicians have long suspected that more potent medications, such as antiretroviral therapy, are associated with adverse effects that are not typically characteristic of HIV, including changes in body fat, elevated cholesterol and triglycerides, and metabolic syndrome.

As HIV medications continue to improve and physicians closely follow the associated adverse effects — putting HIV-infected patients on cholesterol-lowering drugs along with antiretroviral therapy (ART), for example — HIV has become more of a chronic disease and less of a death sentence in recent years.

Ashok Balasubramanyam, MD

Ashok Balasubramanyam, MD, investigates atherosclerotic markers in HIV-infected patients who take ART.

Photo credit: Agapito Sanchez,
Baylor College of Medicine

“Patients with HIV are taking very effective ART and are living longer. As a result, they are becoming susceptible to all of the typical metabolic and cardiovascular diseases that the aging general population faces,” Kevin Yarasheski, PhD, professor of medicine, cell biology and physiology at Washington University School of Medicine, St. Louis, told Endocrine Today.

“HIV infection itself is a risk factor for endocrine, metabolic and CV disorders,” he said. “In addition, some HIV medications are known endocrine disruptors or have been associated with poor glucose control, dyslipidemia, obesity and bone loss.”

Still, disagreement exists about when and how to screen for these risk factors in patients with HIV, what the optimal measures are to use for screening, and when and what ART regimens to use. Depending on prior risk factors, therapy can vary from patient to patient.

Association with diabetes

Diabetes continues to be studied as a risk factor in patients with HIV who are taking ART.

Recent data published in the October issue of Diabetes Care suggest that systemic inflammation may contribute to diabetes pathogenesis in people with HIV. Todd T. Brown, MD, and colleagues reviewed the large ACTG Longitudinal Linked Randomized Trials (ALLRT) database for patients with HIV taking ART. They selected 55 HIV-infected adults who were ART naïve before ALLRT enrollment who subsequently developed diabetes; these patients were compared with 55 HIV-infected adults who did not develop diabetes.

According to the results, inflammatory markers at 48 weeks after ART initiation were associated with an increased risk for diabetes. Median levels for all inflammatory biomarkers examined, except for high-sensitivity C-reactive protein, decreased from baseline to week 48. Patients with higher levels of high-sensitivity C-reactive protein, soluble tumor necrosis factor (TNF)-alpha receptor-1 and soluble TNF-alpha receptor-2 at 48 weeks had an increased odds of subsequent diabetes, after adjustment.

“One of the big issues in HIV in the Western world is not so much the care of the HIV — because drugs today are very effective at controlling the virus — but whether the residual inflammation, even with good ART, is detrimental,” Brown said in an interview. “We know from the general population that baseline inflammation is associated with incident diabetes, independent of body composition.”

Brown said there seems to be a “possible and plausible mechanism” for inflammation in HIV-infected patients taking ART. Other research is currently focusing on mouse models to determine the role of inflammation of TNF-alpha and interleukin-6 in the pathogenesis of diabetes in this patient population, he noted.

The Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) Study, which was published in Diabetes Care in 2008, found that combination ART with stavudine and zidovudine was significantly associated with diabetes, after adjustment for risks factors for diabetes and lipids. Adjustment for lipodystrophy did not modify this relationship, which suggests that the two thymidine analogs “probably directly contribute to insulin resistance, potentially through mitochondrial toxicity,” researchers concluded.

Fast Facts

The prospective, observational study included more than 33,000 HIV-infected patients. During follow-up, diabetes was diagnosed in 744 patients at an incidence rate of 5.72 per 1,000 person-years. The incidence of diabetes increased with cumulative exposure to combination ART. The strongest relationship with diabetes was exposure to stavudine; exposures to zidovudine and didanosine were also associated with an increased risk for diabetes.

“The pathogenesis of diabetes and insulin resistance in HIV is complicated,” Brown said. “Some ART causes insulin resistance either through direct mechanisms or indirectly through effects on body composition. Then, there are the traditional risk factors for diabetes, which may or may not be more common in the HIV population.”

Association with metabolic disease

Recent research has also shown that HIV is associated with severe metabolic complications, including lipodystrophy, dyslipidemia and insulin resistance, a clustering of risk factors that is similar to that of metabolic syndrome. A better understanding of the links between metabolic diseases and morbidities in patients with HIV who are taking ART may allow more effective clinical management of this patient population, experts said.

Kevin Yarasheski, PhD
Kevin Yarasheski

Yarasheski and colleagues at Washington University School of Medicine in St. Louis found that the occurrence of metabolic syndrome was similar in a group of HIV-infected patients taking ART compared with the general population: 25.5% vs. 26.5%. The results of their 2007 study comparing 471 patients with HIV and a matched cohort of the National Health and Nutrition Examination Study population without HIV were published in Clinical Infectious Diseases. The researchers also found that patients with HIV were more likely to have diabetes and a high BMI and to be older and white. Another important risk factor for metabolic syndrome was a high CD4 cell count.

“Overall, the HIV-infected people tended to weigh less and have lower HDL levels, higher triglycerides and lower glucose levels,” Yarasheski said in a press release. “But, it didn’t seem that HIV therapy was influencing their risk for metabolic syndrome as much as the more traditional risk factors that everyone faces.” Type or duration of HIV therapy was not found to be an independent risk factor for metabolic syndrome.

The researchers said that one strength of the study was that they examined these findings by race. Among HIV-infected black women, the rate of obesity was nearly 45%. With a rate that high, they said it is likely time to start investigating the efficacy of exercise and other weight-loss methods in patients with HIV.

Ward and collegues published data in 2007 showing that substantial progression to metabolic syndrome occurs within 3 years after ART initiation in patients with HIV. Of 881 HIV-infected adults examined, 8.5% had baseline metabolic syndrome using the Adult Treatment Panel III (ATP-III) criteria and 7.8% had the syndrome using the International Diabetes Federation (IDF) criteria. More patients (n=234) progressed to metabolic syndrome during follow-up using the ATP-III definition vs. IDF definition (n=178). Researchers also found that both metabolic syndrome at baseline and incident metabolic syndrome were associated with increased risk for CVD and type 2 diabetes.


Studies of the etiology of HIV are helping physicians understand that CV risk in this patient population is “in part due to traditional risk factors and in part due to the virus,” Steven Grinspoon, MD, director of the program in nutritional metabolism at Massachusetts General Hospital in Boston and professor of medicine at Harvard Medical School, said in an interview.

Steven Grinspoon, MD
Steven Grinspoon

“A number of studies suggest that CV events are increased approximately twofold among patients with HIV,” Grinspoon said in an interview. “The relative increases may be more significant among older HIV patients.”

Grinspoon noted findings from Triant and colleagues, who observed a 1.75 (95% CI, 1.51-2.02) greater risk for acute myocardial infarction in a cohort of 3,851 patients with HIV compared with 1,044,589 seronegative individuals after adjusting for age, sex, race, hypertension, diabetes and dyslipidemia.

According to Grinspoon, “subclinical inflammation and endothelial dysfunction associated with the infection may contribute to the increased risk.”

Ashok Balasubramanyam, MD, estimates that “somewhere between 60% and 90% of all patients with HIV who are taking ART end up with dyslipidemia and insulin resistance.

“The basis of this atherosclerotic risk appears to be mediated tremendously by insulin resistance and vascular inflammation,” said Balasubramanyam, professor of medicine in the division of diabetes, endocrinology at metabolism at Baylor College of Medicine. “The onset of metabolic derangement is rapid,” he said, noting that studies show the average time from onset of HIV treatment to increase in atherosclerosis markers is about 1 year. The occurrence of CV events, such as myocardial infarction, has been reported within a few years of onset of treatment.

“It is an accelerated form of atherosclerosis happening in younger people — even people who do not have traditional risk factors.

Grinspoon added that atherosclerotic plaques may be associated with age and Framingham risk score, as well as with the ratio of CD4 and CD8 cell counts and longer duration of HIV.

“The length of time an individual has HIV may be a surrogate for longstanding clinical inflammation,” he said.

Whether the ART is to blame has been a subject of research.

Findings from the Strategies for Management of Anti-Retroviral Therapy (SMART) study indicated that interruptions in ART increased risk for CVD. SMART results also suggested that abacavir use was associated with a fourfold increase in MI incidence.

Priscilla Hsue, MD
Priscilla Hsue

“It was initially thought that ART was associated with increased CV risk,” Priscilla Hsue, MD, assistant professor of medicine at the University of California, said in an interview. “There was definitive evidence of protease inhibitors (PIs) associated with increased CV risk. However, the SMART study showed that in the short term, controlling HIV-related inflammation with ART is good.”

George Behrens, MD, PhD, an assistant professor of T-cell immunology at Hanover Medical School in Germany, said during a presentation at the 2010 International AIDS Conference that most HIV therapies are connected to modest increases in cholesterol, some increases in triglycerides and have a negligible effect on HDL levels.

“Having said that, we all know that this is oversimplified because certain drugs — non-nucleoside reverse transcriptase inhibitors (NNRTIs), for instance — have a more beneficial effect on HDL, whereas PIs differ in their effects on triglycerides, and that more recently developed drugs such as CCR5 inhibitors also have less effect at all,” Behrens said.

Hsue agreed that many of the findings on specific ART regimens and HIV risk are inconclusive.

“Each of these studies kind of slices and dices the data differently,” she said. “It is up to clinicians to make their own decisions.”

Carl J. Fichtenbaum, MD, of the division of infectious diseases at the University of Cincinnati College of Medicine, said most PIs alter lipids significantly.

Carl J. Fichtenbaum, MD
Carl J. Fichtenbaum

“Lopinavir, ritonavir, and other commonly prescribed PIs, impact the lipids, which in turn affects CV risk,” he said. “Ritonavir is of particular concern because most PI regimens now have a small boost of it. However, it is difficult to determine the impact it is having and to study it because it is used so commonly and in combination with other drugs.”

Switching the cocktail of HIV drugs has been marginally successful, but Balasubramanyam said that drugs are probably not the whole reason for the increase in these complications. One approach in the treatment armamentarium has been to add anti-lipid drugs or insulin-sensitizing drugs such as metformin to ART. The response to some standard lipid-lowering drugs was “less satisfactory” than in the general non-HIV population, he said. Balasubramanyam cautioned adding statins to ART, with the exception of pravastatin or low-dose atorvastatin. The bottom line is, “we do not know what the best cocktail of lipid drugs and glucose-lowering drugs is when added to ART,” he said. “We are very intrigued by the possibility that maybe the HIV itself or the immune response to the virus or treatment could be a major trigger in causing the atherosclerosis and diabetes.

“A clever approach in the long run will be a combination of preventive factors, anti-lipid drugs and effective methods to attack chronic immune activation,” Balasubramanyam said.

Fichtenbaum suggested that pharmaceutical companies are not studying the long-term complications of the medications they produce.

“It is challenging to engage pharmaceutical companies and obtain NIH funding to study long-term complications because there is a limited pot of money and competing priorities,” he said. “It is generally not in the strategic interests of pharmaceutical companies to study complications of the medications they produce. Also, the structure of the NIH into different, somewhat independent funding agencies makes it challenging to study longer-term complications.”

Fichtenbaum cited the example of how the National Institute of Allergy and Infectious Diseases generally funds projects that are associated with infections and the National Heart, Lung, and Blood Institute deal with those related to CVD. “Collaboration between institutes is needed to solve complex problems with support from the pharmaceutical industry,” he said.

One basic problem with ART and the CV system is simply the pill burden, according to Fichtenbaum. He noted that the drugs are difficult to process for the kidneys, exacerbating the associations between metabolic and CV diseases.

Current state of complications

Some of these complications are less frequent in Western countries, as drug therapy has changed, according to Katherine Samaras, MD, associate professor and head of the Diabetes and Obesity Clinical Group at Garvan Institute of Medical Research, Australia.

Katherine Samaras, MD
Katherine Samaras

“In countries where there is access to newer agents within the nucleoside reverse transcriptase inhibitors and NNRTI classes, we see less and less of the disfiguring lipodystrophy,” she told Endocrine Today. “However, these are agents that are being offered first line in poorer nations, such as Africa and South East Asia. Thus, at a global level, these adverse effects will still present significant comorbidities.”

In addition, Samaras noted that physicians are seeing more osteoporosis in patients with HIV, which appears to be related to a number of factors, including duration of ART and specific ART drug classes.

“We should also be mindful of other non-endocrine complications that substantially impact health and well being, including neuropathy; cognitive decline, which is currently under intense investigation; and increased risk for malignancies such as non-Hodgkin’s lymphoma, anal cancer or premalignant conditions,” she said.

For clinicians

Hsue said that clinicians should be aware that CVD risk is increasing in patients with HIV, and stressed the importance of assessing familial risk factors. She added that the symptoms of CVD can be subtle.

“At our HIV and CV clinic in San Diego, we often get patients referred to us after the cardiac event, but as cardiologists, we want to intervene before the event,” she said. “We want to diagnose them and treat them before they have their heart attack.”

A key diagnostic issue in recent research surrounds measuring risk for CVD in patients with HIV. Grinspoon said that using the Framingham risk score is “good but not perfect” in HIV populations. “The score may underestimate risk, particularly in smokers, which is obviously a concern given the prevalence of smoking among patients with HIV.”

Grinspoon suggested that an equation similar to the Framingham developed specifically to assess risk in the HIV population may be useful score may assist clinicians in assessing CV risk. “It may be helpful to have an equation in which the traditional and nontraditional inflammatory and immunological risk factors may be included,” he said.

A more controversial issue is how aggressively to use highly sensitive, but expensive and still experimental screening techniques such as coronary angiography, according to Grinspoon. There is no consensus on this approach as yet, but early studies are showing that these techniques can pick up a high prevalence of soft, noncalcified plaque, which may be more vulnerable to rupture, in the HIV population.

Screening for inflammatory markers, which are generally higher in HIV populations, may be beneficial for assessing both CV and renal risk, according to Fichtenbaum. He also said that using specialized imaging techniques such as CT scans to find calcium deposits could help clinicians determine CV risk.

“Inflammatory markers like C-reactive proteins may not necessarily be associated with higher cardiac risk in individuals with HIV,” he said. “These strategies need more research and are not ready to be used every day by patients and clinicians.

“C-reactive proteins may not necessarily be associated with higher cardiac risk,” he said. “And though these strategies have research value, they may not be clinically valuable to the patient.”

Fichtenbaum said there is a lot of research being conducted in these fields, but “the complications and risk factors are many.”

It is necessary to “go back to the science” to try to determine the mechanisms underlying the many complications in patients with HIV taking ART, Balasubramanyam said.

“The approach for endocrinologists is to think of this as a unique endocrine syndrome — not one thing but a mix of things,” he said. “The truth will finally come out that these HIV-related complications are part of a complex disease that involves the virus, immune response to the virus, immune response to the treatment, and perhaps some direct effects of the drugs.”

All aside, the face of the typical HIV patient is changing.

“We now have HIV-infected people who are well into their 70s walking into the clinic with typical age-associated frailties,” Yarasheski said. “We need to start talking across disciplines about common mechanisms and common pathogeneses that may help us discover better ways to treat patients with HIV who have these endocrine, metabolic, CV and inflammatory conditions.” – by Katie Kalvaitis and Rob Volansky

How aggressively should patients with HIV be screened for renal risk?


Active screening should be required for all patients

As more and more people with HIV benefit from effective ART, they are living longer but face new challenges, particularly premature aging. Patients are disproportionately affected by atherosclerotic heart disease and its risk factors, diabetes, hypertension and chronic kidney disease. CKD, either reduced kidney function (estimated glomerular filtration rate less than 60 mL/min/1.73m2) or abnormally elevated urinary protein excretion, is an important predictor of CVD and mortality in HIV. It is generally asymptomatic but treatable. It is relatively rare in younger patients but its prevalence increases with aging. In the United States, kidney disease clusters with health care disparities associated with race, injection drug use and socioeconomic status. Active screening is therefore required for early detection and intervention.

Jonathan A. Winston, MD
Jonathan A. Winston

Infectious Diseases Society of America guidelines published in 2005 remain relevant. Kidney function should be tested using serum creatinine and a routine urinalysis in all HIV-infected patients. Patients at higher than baseline risk for kidney disease because of race, family history, diabetes, hypertension, co-infection or AIDS-defining illness should be tested annually. A routine urinalysis is insensitive to low but epidemiologically important levels of urinary albumin excretion. In patients with diabetes, microalbuminuria predicts future clinically significant declines in kidney function, so screening in these patients must include the urinary albumin-creatinine ratio. Cumulative evidence indicates that microalbuminuria in non-diabetics can predict future CV events. Screening HIV-infected patients already at risk for atherosclerosis, because of hypertension or dyslipidemia, might be worthwhile because its presence could lead to more aggressive CV risk factor modification and more liberal use of angiotensin inhibitors in hypertensives.

Jonathan A. Winston, MD, is professor of medicine in the department of renal medicine at Mount Sinai School of Medicine.


Screening should be as aggressive as for patients with diabetes

After many years of research, we can definitively point to all of the papers that complete our ability to shift the paradigm that we use in people with diabetes to one for use in people with HIV. It is ingrained in every internist that patients with diabetes should get a urine test twice a year. The main reason for this is that small amounts of protein predict larger amounts of protein, and both predict CV morbidity and mortality as well as progression to end-stage renal disease. Just like in diabetes, the ability to predict CV morbidity and mortality is even stronger than the ability to predict the progression to dialysis. Because we have all of the links that we had in diabetes that drove us for decades to test urine twice yearly, we now have similar evidence to use that strategy in patients with HIV.

Lynda Anne Szczech, MD
Lynda Anne Szczech

There have been concerns that urine testing is labor intensive, particularly in light of the fact that only a handful of patients will progress to dialysis. The important point here is that people with kidney disease or urinary abnormalities are more likely to die than to progress to dialysis. Of course it would be optimal to prevent progression of kidney disease; however, if we can’t prevent kidney disease, then keeping people alive long enough for their kidney disease to progress option could still be considered a “win.”

The whole idea of screening is to treat the disease. That said, protein in the urine can be considered a sign of kidney disease but needs to be recognized that it is an even stronger sign of cardiac disease. Protein in the urine is an even more potent marker than cholesterol. Every patient who has had a cholesterol test should also have their urine protein done twice as frequently, if not more frequently. It is cheap, easy and the best way to identify a person who is a heart attack waiting to happen.

Lynda Anne Szczech, MD, is associate professor in the division of nephrology at Duke University Medical Center.

For more information:

  • Brown TT. Diabetes Care. 2010;33:2244-2249.
  • De Wit S. Diabetes Care. 2008;31:1224-1229.
  • El-Sadr WM. N Engl J Med. 2006;355:2283-2296.
  • Lucas GM. JAIDS. 2010. doi: 10.1097/QAI.0b013e3181e8d5a8.
  • Nicolau Laparra MC. Nefrologia. 2010;30:420-426.
  • Triant VA. J Clin Endocrinol Metab. 2007;92:2506-2512.
  • Wand H. Aids. 2007;18:2445-2453.