Updates

by Carol A. Kemper

Multivitamins delay progression to AIDS

Source: Fawzi WW, et al. N Engl J Med. 2004;351:23-32.

An 8-year study conducted in Dar es Salaam, Tanzania, has found that daily multivitamins (vitamins B, C and E) can delay progression of HIV disease in African women. Between 1995-1997, a total of 1078 pregnant women with HIV were enrolled in this double-blind, placebo-controlled trail initially comparing MVI (vitamin B,C, and E), vitamin A, the combination of MVI and vitamin A, and placebo. None of the women had access to antiretroviral therapy.

The study was later modified so that the women received MVI during all pregnancies, and then after delivering they resumed their previously assigned blinded therapy. The women also received standard doses of folic acid and iron during their pregnancies. Vitamin A was later dropped from the study when safety analyses suggested it increased the risk of neonatal transmission of HIV. During a median of 71 months of follow-up, 67 of 271 women who received MVI developed World Health Organization (WHO) Stage 4 AIDS or died (a primary outcome), compared with 83 of 267 women who received placebo (24.7% vs 31.1%, relative risk 0.71; P = .04).

MVI also resulted in a statistically significant slower progression to (WHO) Stage 3 disease, as well as significantly higher CD4 counts and significantly lower viral loads. MVI also significantly reduced the frequency of a number of other complications, including oral ulcers, angular cheilitis, difficult or painful swallowing, dysentery, rash, fatigue, and acute upper respiratory infection. In comparison, women assigned to receive vitamin A alone had no apparent benefit, and may have had an increased risk of certain symptoms, such as angular cheilitis and difficult or painful swallowing. The addition of vitamin A to MVI actually appeared to diminish the effect of MVI—suggesting that the administration of vitamin A should be avoided in patients with HIV.

While most of these women were poor, they were not malnourished but had fairly reasonable diets for Africans. For this reason, Fawzi and colleagues believe that the additive benefit of MVI could be measured. In other words, while the effect of MVI may not be apparent in HIV-infected individuals living in theUnited States, who by comparison have diets rich in nutrients, supplemental MVI made enough of a difference in these womens’ African diet to help maintain gastrointestinal mucosal integrity, local and cellular immunity, and cytokine responses. In addition, both vitamin C and E are excellent anti-oxidants. Given an estimated retail cost of $15 per person each year, the use of MVI may be a relatively inexpensive stopgap measure for HIV-infected individuals living in third world countries with limited access to anti-retrovirals.

Tenofovir Decreases Atazanavir Levels

Source: Taburet AM, et al. Antimicrob Agents Chemother. 2004;48:2091-2096.

Atazanavir (ATV), an azapeptide protease inhibitor with a distinct resistance profile, functions as a CYP3A substrate. Hence, ATV concentrations in the blood stream can readily be boosted with low-dose ritonavir. In healthy volunteers, the addition of 100 mg ritonavir daily to 300 mg ATV daily leads to a 3- to 4-fold increase in the area under the curve (AUC) in the bloodstream, and the half-life of ATV increases from 6.5 hours to 15-18 hours. However, a number of factors may lower ATV concentrations, including autoinduction, diminished absorption in the presence of a fatty meal, and the co-administration of tenofovir (TNF).

Following a 2-week course of ATV plus RTV, 11 HIV-infected patients additionally received 300 mg TNF per day. Peak concentrations, AUC, and t1/2 were measured at week 2 and week 6. With the addition of TNF, there was a significant decrease in the 24-hour AUC for ATV (~27% decrease), and both the Cmin and Cmax values dropped by about one-third, although there was significant individual variability. The half-life remained unchanged, suggesting that TNF had no effect on ATV clearance. Although the

exact mechanism of the interaction between the 2 drugs remains unclear, Taburet and colleagues speculate that absorption of ATV could be impaired in the gut by the co-administration of TNF, possibly through the induction of P-glycoprotein transport mechanisms. Although staggering the administration of the 2 agents may diminish this effect, further information is needed before these 2 drugs should commonly be used together.

Breakthrough Chickenpox in Vaccinated Kids

Source: Vasquez M, et al. JAMA. 2004,291:851-855.

Breakthrough varicella in previously vaccinated kids is not uncommon. The occurrence of several recent outbreaks of chickenpox in heavily vaccinated pre-school groups led Vasquez and colleagues to re-evaluate the effectiveness of varicella vaccine in a large group of children who developed chickenpox (matched by age and site of clinical care), compared to children without chickenpox. In this case-control study, 339 children with chickenpox were compared with 669 control subjects who had not had chickenpox (aged 13 months or older). All cases were confirmed by PCR of active lesions.

Overall, 36% of case patients had a history of varicella vaccination compared with 70% of uninfected controls. Chickenpox was significantly more likely to be more severe in unvaccinated children; 45% vs 87% of unvaccinated vs vaccinated children with chickenpox had mild disease. Interestingly, the rash was more likely to be predominately vesicular in unvaccinated children compared with those who had been vaccinated (58% vs 30%). Overall, the vaccine was 97% effective during the first year post-vaccination, but decreased to 86% by the second year, and to 81% by 7-8 years following vaccination. Vaccine efficacy was significantly lower if the vaccine was administered before 15 months of age (73%), although the earlier administration of vaccine still protected against more severe infection.

These data provides continuing evidence for the overall benefit of varicella vaccine—if not so much for the prevention of chickenpox but for a reduction in the risk of more severe disease. If the objective is to reduce rates of breakthrough infection, then the administration of vaccine should be delayed until at least 15 months of age, and consideration should be given to a booster dose later in life. However, shifting the timing of vaccination to slightly older babies leaves them vulnerable to infection for those additional few months, and some children may not return for vaccination. Probably the best bet is to offer vacation to kids (ie, moms) as they present for care at about >1 years of age, with a good explanation that vaccination does not prevent infection: ~14-27% of kids still get chickenpox but they are much more likely to have milder disease.

Find the VRSA in the MRSA: Get an Etest!

Source: MMWR Morb Mortal Wkly Rep. 2004;53(15):322-323.

Routine automated susceptibility testing of methicillin resistant Staphylococcus aureus (MRSA) may fail to detect vancomycin resistance (MIC >32 µ/mL). The Centers for Disease Control & Prevention is alerting physicians and micro labs, following discovery of a third case of vancomcyin-resistant S. aureus (VRSA) infection in the United States, that initially escaped detection. The patient was a resident of a long-term care facility, who developed a urinary tract infection with MRSA. Susceptibility testing by MicroScan (Dade Behring, Deerfield, IL) showed a vancomycin MIC = 4 µ/mL. Further testing by Etest (AB Biodisk North America, Inc, Piscataway, NJ) demonstrated vancomycin resistance with an MIC > 256 µ/mL.

After first being tested by the New York Public Health Department, the isolate was forwarded to the CDC, where MicroScan and Vitek (bioMerieux, Hazelwood, MI) methods failed to detect vancomycin resistance. Using NCCLS guidelines, broth microdilution testing confirmed vancomycin resistance with an MIC = 64 µ/mL. Further tests confirmed that the isolate contained both mecA and VanA genes—encoding methicillin and vancomycin resistance, respectively, but was unrelated to the 2 earlier VRSA isolates identified in Michigan and Pennsylvania last year. The isolate was also susceptible to linezolid, quinopristine-dalfopristine, minocycline, rifampin, trimethoprim-sulfamethoxazole, and chloramphenicol.

There is increased concern that commonly employed automated methods for susceptibility testing in microbiology laboratories may fail to detect vancomycin intermediate and resistant MRSA strains. The CDC recommends that micro labs should include a non-automated MIC method (broth microdilution,agar dilution, or agar-gradient diffusion) when setting up S. aureus isolates, especially MRSA isolates, for automated testing.

Fortunately, aggressive screening of family members and fellow patients and employees of the LTCF failed to identify any secondary cases.

Carol A. Kemper, MD, FACP, Clinical Associate Professor of Medicine, Stanford University, Division of Infectious Diseases; Santa Clara Valley Medical Center, Section Editor, Updates, Section Editor, HIV.