HIV JournalView: September 2012

David Wohl, M.D.

We tell our patients that HIV has become like other chronic diseases -- and, largely, we believe it. These patients are aging along with us, and clinic visits have become more like routine checkups with continued praise for undetectable viral loads and admonishments to shed pounds. Even our scientific literature has become more mundane; it is less preoccupied with finding silver bullets than with developing strategies to help people live well with their virus into their golden years.

There are exceptions, of course. Developers of an HIV vaccine and developers of a cure are laboring on breakthroughs, and prevention research has yielded interesting, if not provocative, discoveries. But with a cornucopia of antiretrovirals, superb efforts to monitor care delivery, and a deep understanding of who is at risk for HIV infection, we know what we need to do until such dreams become realities.

Below are examples of recent reports that help us stay the course until we have a vaccine and cure. They include studies that take stock of our successes, flash caution about our medications, and point to lurking problems that challenge the prevention and treatment of HIV.

Table of Contents

NA-ACCORD Makes Us Feel Good

Keri N. Althoff et al. Ann Intern Med 2012;157:325-335. Read the abstract.

The most compelling reason to include this article in our latest JournalView installment will be best appreciated by fans of A. A. Milne, Shel Silverstein and Mary GrandPré. Each of these artists gave us a common visual reference for the story being told, whether it was about a paunchy bear with a honey addiction, a flatulent ogre with an attitude or a resilient young wizard with a flashy forehead scar. It is difficult to imagine these books without the drawings that intermittently grace their pages.

Similarly, this article -- the latest "how are we doing?" from the NA-ACCORD cohort of HIV-infected patients in care -- is pretty cool, but it is the pictures that make it special. There is the map, all in shades of blue (no red states!), showing the proportion of people with HIV infection who are engaged in medical care in different U.S. states. (You go, Washington, Louisiana and Tennessee!) Then, there are the bar graphs of changes in antiretroviral therapy use over time (I swoon for horizontal bar graphs, don't you?), not to mention a multilayered, 4-D graphic of temporal trends in viral suppression. Marvelous.

We could show these images here, but our lawyers won't let us. Happily, there are also some words in the study worth looking at. The short version goes something like this:

  • The proportion of cohort participants prescribed HAART increased from 74% in 2000 to 83% in 2008.
  • During this time, the percentage of patients with a suppressed viral load increased from 54% to 81%.
  • The CD4+ cell count prior to death, among those who died, increased over this eight-year study period from 60 to 204 cells/mm3.
  • Lastly, to place this all in perspective, the characteristics of the 45,000 or so HIV-infected patients in the U.S. under the watchful eye of this cohort from 2000 to 2008 are very similar to those of the more than 650,000 included in the U.S. Centers for Disease Control and Prevention (CDC)'s 40-state HIV surveillance system (with the exception of fewer injection drug users in NA-ACCORD), suggesting that the results bear a similar representation to that of all people living with HIV in the U.S.

These data will hopefully be mined for more gems. For example, other studies have shown disparities in HIV care outcomes, such as in HAART prescriptions and viral suppression, by race in the U.S. Similarly, retention in care is a challenge and understanding who is at greatest risk of dropping out can lead to interventions (see below regarding the possible role of pain). These should be easy to examine in this cohort.

Meanwhile, I can't help but feel great after reading an NA-ACCORD article. The cohort study is one of the best at showing how the epidemic is evolving and what we are collectively accomplishing. And there are the pretty pictures.

Pain Makes Our Patients Skip Appointments

Jessica S. Merlin et al. J Acquir Immune Defic Syndr 2012;61:164-170. Read the abstract.

Once we started asking our patients to point out which bald-headed face represented their level of discomfort -- choices ranged from the euphoric to the tear-shedding despondent -- we came to feel their pain. Now practically a vital sign, pain scales are supposed to help us discover and address suffering.

Of course, the validity of such scales in those with HIV is not clear. We all have had patients who respond that they have 10 out of 10 pain (that is the worst pain imaginable), but are smiling and moving about without any outward display of distress.

But pain is complex, and it can be life-sapping. Patients living with HIV may be more likely to experience pain as a consequence of the virus and its therapies, as in the case of peripheral neuropathy, or due to comorbid conditions far removed from their infection.

A study conducted at the University of Alabama-Birmingham examined the prevalence of pain and its impact on clinic visits (especially no-shows), adherence to HIV medications and viral suppression. The study cohort included 1,521 patients (mean age 44 years; most were non-white, male, uninsured or publically insured; and 64% had a CD4+ cell count >350 cells/mm3). A third of the patients reported pain (more than 70% recorded moderate or extreme pain at the initial point of data collection), a quarter had mood disorders and 10% reported substance abuse.

In multivariate modeling, the presence of pain increased the odds of a subsequent clinic no-show, but only in those without substance abuse (OR 1.5; 95% CI, 1.1-1.9). Pain actually reduced the odds of no-shows for those with substance abuse (OR 0.5; 95% CI, 0.2- 0.9, P for interaction = 0.0022). Similarly, substance abuse increased the odds of a no-show visit only in those without pain, while for patients with pain, substance abuse had no effect on the odds of a no-show visit.

Other factors that were associated with a no-show were being a non-white female, a non-white male or a white female; a lack of private insurance; and a CD4+ count above 200 cells/mm3. Giving credence to the whole age-equals-wisdom thing, increasing age was inversely associated with not showing up for a visit. Pain was not correlated with antiretroviral adherence or with virologic outcomes.

These results are interesting, if not somewhat validating of clinical experience. Pain among those with substance abuse seemed to be a motivator for medical visit adherence. (OK, most of you are probably thinking, "Duh," and perhaps this is explained by the fact that pain medications such as opioids are prescribed during clinic visits -- or at least the patients hope they will be.) For those without substance abuse, the link between pain and missing visits is not as intuitive. Perhaps these folks were too sick to come in or were frustrated that their pain was not being adequately addressed.

Given the importance of retention in care to maintaining the benefits of treatment (in terms of personal well-being and transmissibility), the finding that better pain management may keep people in care is significant. As the needs of our patients expand, we HIV providers have to be many things to many people. Adding "pain specialist" to that list can be tricky, but may help our patients in more ways than we expected.

Atazanavir Stones: The Gallbladder

Soloniaina Rakotondravelo et al. Clin Infect Dis 2012;55(9):1270-72. Read the abstract.

The crystallization of atazanavir (Reyataz) in the urine, which can lead to kidney stones, has been well described. In addition to hyperbilirubinemia, this is a major, albeit infrequent, adverse effect of this popular protease inhibitor. Unlike indinavir (Crixivan), which had patients perpetually on the knife's edge of nephrolithiasis, atazanavir has most clinicians only occasionally encountering patients toting telltale burnt sienna stones in a jar for their doctor visit show and tell.

But now comes a new wrinkle. A report from France raises concern that atazanavir can also form stones in the gallbladder. In this case series, 14 cases of cholelithiasis developing in patients receiving atazanavir were described. (The median duration of atazanavir use was 41 months [range 1-90], 12 patients were also on ritonavir [Norvir] and seven patients were hepatitis C coinfected.)

Yes, they were on atazanavir. Yes, they had gallstones. Yes, atazanavir has been known to accrete in the kidneys. J'accuse! But wait: Stones happen. Was atazanavir really the culprit?

The sleuthing clinicians rounded up gallstones in 11 of the cases and sent them to the lab for spectrometry analysis. Voilà! In eight cases, the stones had atazanavir in high concentrations; in the others, the calculus was composed of the more typical calcium bilirubinate. While a role for atazanavir in all 14 cases was not confirmed, the finding of chunks of crystalline atazanavir in the gallbladder of a few of these patients is damning.

The mechanism by which atazanavir may lead to cholelithiasis remains a mystery, but it could involve precipitation of the drug in the bile. Or perhaps atazanavir promotes an increase in the concentration of the components of typical gallstones. We may never know how atazanavir does it. Meanwhile, gallbladder colic should be considered in any patient with significant right-upper abdominal pain -- and, in the case of those on atazanavir, the drug should be a prime suspect.

Atazanavir Stones: The Kidney

Yohei Hamada et al. Clin Infect Dis 2012;55(9):1262-9. Read the abstract.

Now that atazanavir-studded gallstones are the newest thing from France, good old-fashioned clumps of the stuff in the kidney are passé. However, although not believed to be very common, the actual incidence of atazanavir kidney stones has not been clear, leaving a gap in knowledge worth filling.

In Japan, the records of 1,240 HIV-infected patients starting a ritonavir-boosted protease inhibitor were examined to look at the rate of nephrolithiasis during treatment. Of the cohort, only 37.5% were prescribed atazanavir, yet of the 35 cases of nephrolithiasis recorded, 31 occurred in those taking that very protease inhibitor -- an incidence rate of 23.7 cases per 1,000 person-years.

The median time from initiation of atazanavir/ritonavir to diagnosis of kidney stone was 24.5 months (range: 14.7-34.6 months). Only four patients receiving other protease inhibitors experienced kidney stones (2.2 cases per 1,000 person-years). Of the 18 who forged ahead with atazanavir despite nephrolithiasis, six had a recurrence, while none of those who stopped the drug had another bout of kidney stones. Multivariable analysis failed to find factors other than atazanavir exposure that increased risk for nephrolithiasis.

This study was conducted among an Asian patient cohort and was retrospective; however, the 10-fold higher risk of kidney stones with atazanavir compared to other protease inhibitors fits, and the finding allows for a decent estimation of the rate at which kidney stones can be expected in those prescribed atazanavir.

Role of Anal Sex in HIV Risk for Women in Chicago

Britt S. Livak et al. Open AIDS J 2012; 6(Suppl 1: M6):142-8. Read the abstract.

The role of anal intercourse in HIV transmission among heterosexuals is understudied, despite evidence that many women, including those most at risk for HIV acquisition, engage in anal sex. According to one recent study, approximately 9% of women in the general U.S. population had anal sex within the past 12 months, and in a national probability sample, a third of the women aged 20 to 44 reported ever having anal sex.

Other data find that anal sex, particularly among injection drug users, was associated with additional risks for sexually transmitted infections (STIs). Less is known about specific sexual practices, such as anal sex, among non-injecting women, such as those living in inner cities in poverty.

This study from the Chicago Department of Health piggybacks on the recruitment of participants in a CDC study of people at risk for HIV in 21 cities across the U.S. To make sense of the study results, it is important to understand who exactly was recruited. The study involves identification of geographic areas considered higher-risk for heterosexual HIV transmission based on HIV prevalence and poverty level. People frequenting venues (convenience stores, laundromats, street corridors, etc.) within a selection of these areas were recruited and surveyed.

Of the 1,101 people approached, 851 consented; 48% were women. The mean age of women participating in the study was 31.7 years; the majority were black (82%) and had an annual income less than $10,000 (81%). Over the past year, 15% were homeless and 11% had been arrested and detained for over 24 hours. Excluding five who reported injection drug use, 87% reported unprotected vaginal sex and 17% reported anal sex (for 86% of those women, the anal sex was unprotected) in the past 12 months.

Rapid HIV testing was accepted by 381 (94%) of the women in the study; seven (1.86%) were positive, with no major difference in positivity rates between those with and without a history of anal sex.

In a multivariable analysis of factors associated with anal sex (adjusted for age and having had unprotected vaginal intercourse in the past 12 months), three variables were independently correlated with anal intercourse in the past 12 months: having three or more sex partners in the past 12 months (OR 3.27, 95% CI, 1.53-6.99), self-reported STI diagnosis in the past 12 months (OR 2.13, 95% CI, 1.06-4.26) and having had vaginal or anal intercourse for the first time before the age of 15 (OR 2.23, 95% CI, 1.28-3.89).

The prevalence of these and other STI risk factors in this sample -- including concurrency, prior diagnosis with STIs, illicit substance abuse and alcohol abuse -- was striking. Moreover, almost 90% of participants reported they had neither been exposed to nor utilized HIV prevention services over the previous year.

Importantly, the authors draw attention not only to the potential use of their data to profile women who may have higher risk of HIV acquisition via anal sex, but also to the structural factors they discovered that likely promote HIV/STI transmission in these women. Such structural factors include endemic homelessness, unemployment, low education attainment and incarceration.

Remedies for these social ills are a challenge to develop. Unlike interventions that target individual behavior, those addressing the structures that exist in a community may be difficult to apply and test. However, a greater obstacle to minimizing the effects of poverty, discrimination and similar forces is a failure to acknowledge their existence and a lack of willingness to take meaningful action to mitigate their toxic effects.

The current political debate on the role of government, which is uniquely able to intervene on a scale necessary to affect significant change, makes it clear that many in the U.S. do not believe we should be righting these wrongs. Until the scales tip more in favor of such action, we will continue to see grim reports such as this reminding us of the cost of inaction.

Echocardiography Misses Pulmonary Artery Hypertension in a Third of Cases

Van N. Selby et al. AIDS 2012, 26:1967-9. Read the abstract.

Pulmonary artery hypertension, a dangerous complication of HIV infection, is challenging to diagnose. Typically, the workup starts with a doppler echocardiogram. However, in a study from San Francisco, echocardiography was found to be insensitive for detection of pulmonary artery hypertension when compared to the gold standard of right heart catheterization.

The study was conducted among a cohort of 422 HIV-infected patients who underwent a research cardiac echo. Of these, 129 had an estimated pulmonary artery systolic pressure of 30 mmHg or more (an established upper limit of normal). They were all offered right heart catheterization; 76 accepted (80% male, median age 52 years).

While the pulmonary artery pressures estimated by the doppler echocardiography and the heart catheterization were generally correlated, the echo tended to overestimate pressures overall, but underestimate pressures when values were high. This meant that for 14 patients who were determined by right heart catheterization to have pulmonary artery hypertension (pressure >35 mmHg), five were missed by echocardiography.

These concerning results mirror those from studies of uninfected patients. HIV clinicians should take note of the limits of echocardiography and appreciate that when pulmonary artery hypertension is on the differential, a negative echo does not rule out the diagnosis. Only heart catheterization can be conclusive.