IDSA News - October-November 2013 (Plain Text Version)
In this issue:
IDSA Journal Club
In this feature, a panel of IDSA members identifies and critiques important new studies in the current literature that have a significant impact on the practice of infectious diseases medicine.
Click here for the previous edition of Journal Club. For a review of other recent research in the infectious diseases literature, see “In the Literature,” by Stanley Deresinski, MD, in each issue of Clinical Infectious Diseases.
Reviewed by Terri Stillwell, MD
Despite some indications of a reduction in methicillin-resistant Staphylococcus aureus (MRSA) infections in adults, questions remain regarding such trends among pediatric patients. Two recent studies in Pediatrics address the continued importance of MRSA infections in children, looking at national trends and the role of blood cultures as part of routine diagnostics.
In the first study, researchers utilized the Active Bacterial Core Surveillance (ABCs) system, a population-based surveillance network covering nine major U.S. cities, to assess national trends of invasive MRSA infections in children. During 2005-2010, 876 cases of pediatric invasive MRSA infections were reported via ABCs. Sources of infection included bloodstream, bone/joint, skin/soft tissue, pneumonia, and other. Of these, 42 percent were categorized as community-associated (CA), 35 percent as hospital-onset (HO), and 23 percent as health care-associated community-onset (HACO) infections.
The authors also estimated national incidence. Although a decreasing trend in HO-MRSA infection among infants 3-89 days old was noted (modeled yearly change -11.3 percent [CI:-19.6 to -2.0 percent]), the highest incidence of invasive infection overall was seen in infants <3 months of age (43.9/100,000 population [CI:29.3-63.9]), with a majority of those being HO-MRSA bloodstream infections in premature neonates. In patients >3 months of age, there were no significant trends in HO- or HACO-MRSA infections. However, CA-MRSA infections increased significantly with a modeled yearly change of 10.2 percent (CI:2.7 to 18.3 percent; p=0.007). Invasive MRSA infections clearly remain an issue within the pediatric population, highlighting the need for continued infection prevention efforts.
With the increasing trend of CA-MRSA infections in pediatric patients, emergency room (ER) visits due to skin and soft tissue infections (SSTI) also seem to be on the rise. In the other study, the authors evaluated the prevalence of bacteremia in otherwise healthy children with SSTI admitted via the ER. Using ICD-9 codes for cellulitis/abscess, researchers reviewed 580 patient charts over a three-year time period. Patients were grouped into uncomplicated (uSSTI, n=482) and complicated (cSSTI, n=98) infections; cSSTI included patients with surgical/traumatic wound infections, those who needed surgical intervention outside of routine incision and drainage, and those with infected ulcers/burns.
Ninety-four percent of uSSTI patients had blood cultures drawn, with zero positive cultures; 82 percent of cSSTI patients had cultures drawn, with 10 positive cultures (six MRSA, three MSSA, one S. pneumoniae; p=<0.001). Furthermore, uSSTI patients whose blood cultures were drawn had an increased length of stay (0.91 days, CI:0.026-1.8 days, p=0.044) when compared to uSSTI patients who did not have blood cultures drawn. The study authors concluded there is a <1 percent chance of bacteremia in the setting of uSSTI and, therefore, routine blood cultures may not be warranted and may, in fact, increase length of stay.
Taken together, these studies underscore that there is more work to be done regarding invasive MRSA infections in children, including infection prevention and appropriate diagnostic strategies.
Studies of the relationship between serum drug levels and clinical outcomes in tuberculosis (TB) have yielded conflicting results. A recent analysis published in The Journal of Infectious Diseases, however, finds an increased risk of poor outcomes among patients with low area under the concentration-time curves (AUCs) for one or more first-line agents.
Researchers studied 142 South African patients ≥16 years old with active pulmonary TB (based upon positive sputum microscopy or culture). None of the baseline isolates carried drug resistance. Sixty-four percent of patients had prior TB; 10 percent had HIV. All were treated with isoniazid, rifampin, pyrazinamide, and ethambutol during a two-month inpatient hospitalization; patients being retreated were also given streptomycin. Drug levels were checked at the end of two months of therapy, after which culture-negative patients transitioned to isoniazid plus rifampin (plus ethambutol for re-treatment patients). All patients were followed for two additional years for treatment failure (defined as smear positivity after six months of treatment), relapse, or death.
Twenty-five percent of patients had poor outcomes, including 19 who relapsed, 15 who died, and two who remained smear-positive at six months. The strongest predictors of poor outcomes were low AUCs for pyrazinamide, rifampin, and isoniazid; ≥91 percent of such patients had a low AUC for at least one of these drugs. Seventeen of 60 patients with a low pyrazinamide or rifampin AUC had a bad outcome, for example, as compared with zero of 64 patients without low AUCs for these drugs (OR = 52).
This study highlights the role that inter-patient pharmacokinetic differences play in suboptimal AUCs and poor clinical outcomes; it also calls to attention the surprisingly pivotal role played by pyrazinamide in sterilizing sputum and clearing smears. While not yet part of routine clinical practice, this work highlights a possible role for therapeutic drug monitoring in the management of TB.
Advances in HIV treatment have made strides against the HIV epidemic, but developing a safe and efficacious preventive vaccine, which would help halt the epidemic, remains an elusive goal. In a study recently published in the New England Journal of Medicine, researchers describe the results of a randomized, placebo-controlled phase IIB study performed by the HIV Vaccine Trials Network (HVTN 505).
In this multicenter trial, 2,504 participants were randomized to receive either placebo or a DNA vaccine encoding multiple HIV proteins followed by a recombinant adenovirus vector boost (DNA/rAd5 vaccine). The trial was performed in the United States and the study population comprised primarily of men who have sex with men. The participants were at substantial risk of HIV infection as all had a history of unprotected insertive or receptive anal sex and 54 percent reported having at least three male partners in the prior three months.
The vaccine was found to have an acceptable side-effect profile and most reactions were mild or moderate. Unfortunately, this study was stopped early for lack of vaccine efficacy. Receipt of the DNA/rAd5 vaccine resulted in no discernible protection from HIV infection as there were 27 infections in the vaccine arm vs. 21 in the placebo arm during the post-vaccination follow-up period (2.8 percent vs. 2.3 percent annual incidence, respectively). There was also no reduction in the HIV viral load set point after HIV acquisition. While the DNA/rAd5 vaccine induced HIV-specific T cells, it did a poor job of inducing the same types of antibody responses associated with the protection from HIV infection in the RV144 HIV vaccine trial. Factors that predicted HIV infection included a history of more than three partners or unprotected receptive anal sex in the three months before study enrollment.
Of the four HIV vaccine strategies that have been tested in large efficacy studies, only one (RV144) has shown a modest protective effect. The disappointing results from the HVTN 505 study again demonstrate the challenges to creating an effective HIV vaccine and the long road ahead for the field.
Prosthetic valve endocarditis (PVE) accounts for 7-25 percent of cases of infective endocarditis and is associated with significant morbidity and mortality. One-third of PVE patients die within a year of the diagnosis. Consensus groups recommend surgical therapy for PVE patients who have heart failure, valve dysfunction, cardiac abscesses, or persistent infection despite maximal medical therapy. However, studies comparing medical vs. surgical therapy for PVE are few and of limited validity because of small sample size and retrospective study designs.
In the September 2013 issue of JAMA Internal Medicine, investigators described the results of a prospective, multicenter, international study that assessed the in-hospital mortality and 1-year mortality rates of PVE patients who had received surgical valve replacement vs. those who had received medical therapy alone. Of 1,025 patients with PVE, 490 (47.8 percent) underwent early cardiac surgery during their initial hospitalization and 535 (52.2 percent) received medical therapy alone.
In the unadjusted analysis and after controlling for treatment selection bias, early surgery was associated with lower in-hospital mortality (HR, 0.44 [95 percent CI, 0.38-0.52]) and 1-year mortality (HR, 0.57 [95 percent CI, 0.49-0.67]) as compared to the medical therapy alone. Notably, after controlling for survivor bias by including surgery as a time-dependent covariate, the in-hospital mortality and 1-year mortality rates for the two groups were similar.
In a post-hoc analysis, patients were divided into quintiles based on the predicted probability of surgery. Patients in the quintiles with the highest predicted probability of surgery had lower mortality rates when they had surgery than did those who had medical therapy alone. However, this subgroup analysis did not adjust for survivor bias.
The study’s strengths include its prospective, multinational design; its size; and the stringent analysis, which controlled for both treatment selection bias and survivor bias for the entire PVE cohort. The findings underscore the gravity of PVE, regardless of surgical intervention. Whether optimal timing of surgery will improve patient outcomes may be decided through future studies.