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.
Predicting Drug-Resistant Organisms in HCAP: Do We Always Need to Go Big?
Reviewed by Nirav Patel, MD
recommend that health care-associated pneumonia (HCAP) be treated with broad spectrum antibiotics to ensure adequate initial therapy. However, the risk of inappropriate therapy must be weighed against the risk of overuse of antibiotics. A multi-centered trial from Japan, described in the October 15 issue of the American Journal of Respiratory and Critical Care Medicine
, builds on prior single-center trials attempting to assess the specific risk factors for resistance for patients presenting with either community-acquired pneumonia (CAP) or HCAP.
This prospective study
collected data from 10 large hospitals, screening 1,742 patients and enrolling 1,413 subjects based on the CAP and HCAP definitions. Aggressive microbiologic evaluation identified the etiologic agent of pneumonia. Endpoints included identification of organisms with resistance to guideline recommended therapy for CAP: ceftriaxone, ampicillin-sulbactam, macrolides, and fluoroquinolones.
Not surprisingly, the HCAP group had higher rates of comorbidities and baseline laboratory abnormalities compared to the CAP group, with a higher mortality (likely driven by inappropriate initial antibiotics).
However, when drug resistance was assessed, the same risk factors were noted for both CAP and HCAP, all with an adjusted OR of ~2-2.5: hospitalization for ≥ two days in the last 90 days, immunosuppression, use of antibiotics within the prior 90 days, use of gastric acid suppression, tube feeding, and non-ambulatory status. In addition, three specific methicillin-resistant Staphylococcus aureus (MRSA) risks were also noted: chronic hemodialysis in the previous 30 days, prior MRSA colonization, and congestive heart failure. The sensitivity of ≥ three risk factors was 47.1 percent; specificity, 90.9 percent; positive predictive value, 49.6 percent; negative predictive value, 90 percent; and accuracy, 83.9 percent. The area under the receiver operating characteristic curve was 0.79, which compared favorably to the prior single centered studies.
Based on the results, a significant number of patients with HCAP could be safely treated with CAP antibiotics without compromising therapy. The authors note that the “type of pneumonia…may not determine the presence…of [drug-resistant pathogens].” Clearly, more specific risk factors are at play, including non-traditional risks such as gastric acid suppression. An accompanying editorial
provides a clear table and flowchart to help clinicians use the results in patient care. Before this approach can become the standard of care, however, a further validation study is needed.
Preventing Hepatic Decompensation in HIV-HCV Coinfected Individuals
Reviewed by Brian R. Wood, MD
Among HIV-infected individuals, hepatitis C virus (HCV) coinfection is common, and HIV accelerates progression to advanced liver disease. This has become a dominant driver of mortality in this population, making strategies to prevent hepatic deterioration of paramount importance. Two recent studies in Clinical Infectious Diseases emphasize the role of early treatment: One advocates early antiretroviral therapy (ART), and the other earlier treatment of HCV.
First, investigators from the United States examined early ART and liver disease outcomes in HIV-HCV coinfected men (median age 47, 61 percent black, 87 percent HCV genotype 1) enrolled in the Veterans Aging Cohort Study Virtual Cohort since 1996.
Participants contributed 46,444 person-years of follow-up. Those who started ART had a significantly lower rate of incident hepatic decompensation (HR 0.72, 95 percent CI 0.54-0.94). ART initiation decreased the likelihood of decompensation by 28-41 percent, on average. Strengths of this study
include the large sample size and the use of marginal structural models, which help eliminate bias for indication. The study is limited because it examined only male veterans and lacks solid data on drug and alcohol use. However, this analysis provides robust evidence for early ART in HIV-HCV coinfected patients and suggests that early ART should be strongly recommended for them.
In the second study
, investigators from Spain performed a retrospective cohort analysis to examine the risk of hepatic decompensation in 892 HIV-HCV coinfected individuals (median age 43, 87 percent male, 88 percent on ART at baseline, 59 percent HCV genotype 1) with stage 3 versus stage 4 fibrosis. The results show higher rates of hepatic decompensation in participants with stage 4 fibrosis as compared to stage 3, as measured by liver biopsy (P=0.023, mean follow-up 5.4 years) or transient elastography (P<0.0001, mean follow-up 2.1 years). However, the authors cite that the rate of decompensation with stage 3 fibrosis is clinically meaningful because decompensations occurred in this group, even within one to three years.
The authors conclude that patients with bridging fibrosis should be prioritized for treatment along with those who have cirrhosis. If resources allow, this seems wise. Otherwise, given the hastened course of liver disease in HIV-HCV coinfection, these individuals require close clinical follow-up and perhaps frequent reassessments of fibrosis. (This is feasible with transient elastography, a modality approved by the Food and Drug Administration earlier this year.)
Together, these studies underscore the importance of early ART and early HCV therapy in coinfected individuals; hopefully new antivirals on the horizon will make HCV treatment for all these patients a reality.
Another Flu Shot Benefit: Meta-Analysis Suggests Reduction in Cardiovascular Events After Vaccination
Reviewed by Manie Beheshti, MD
Data has previously linked vaccination against influenza with a reduction in vascular events, such as stroke and myocardial infarction. However, most of this data is based on observational studies and small randomized clinical trials.
In a recent issue of The Journal of the American Medical Association, investigators performed a comprehensive meta-analysis of all randomized clinical trials of the influenza vaccine that evaluated vascular events as efficacy or safety outcomes. This extensive search of Ovid MEDLINE, EMBASE, and the Cochrane databases dating back to 1946 selected adequate studies assessing vaccine vs. placebo. These studies accounted for approximately 6,700 patients (mean age, 67; 51 percent female; 32 percent with a history of coronary artery disease; mean follow-up time, 8 months).
Pooled results demonstrated a significant decrease in major cardiovascular events in those who received influenza vaccine vs. placebo or control (2.9 percent vs. 4.7 percent; RR: 0.64). The number needed to treat (NNT) to prevent one major adverse cardiovascular event was 58. This beneficial trend was even more pronounced in vaccinated patients with recent acute coronary syndromes (10.2 percent in the vaccinated group vs. 23 percent in the placebo or control group; RR 0.45; NNT: 8).
Although limited by its inherent design as a meta-analysis, this study
provides a thorough review of the available literature and demonstrates a significant reduction in cardiovascular events in those receiving the influenza vaccine. Further, the authors note their intent to go forward with a large randomized trial to better evaluate cardiovascular benefits in the vaccinated.
Since less than half of U.S. adults receive this universally and annually recommended vaccine, this study helps add important information about additional potential benefits linked to vaccination against influenza, particularly for high-risk patients.
Eradication of CRE Gastrointestinal Colonization with Oral Antibiotics
Reviewed by Zeina A. Kanafani, MD, MS
A recent study
in the American Journal of Infection Control
examined whether non-absorbable oral antibiotics may be useful in eradicating gastrointestinal colonization with carbapenem-resistant Enterobacteriaceae
Active surveillance for CRE was conducted through rectal swab cultures. Patients received gentamicin (GM) if their isolate was sensitive to GM but resistant to colistin (COL), or COL if their isolate was resistant to GM but sensitive to COL. If the isolate was sensitive to both drugs, patients were randomized to receive GM alone, COL alone, or a combination. Controls were colonized patients who did not receive treatment. Treatment was continued until eradication, but not exceeding 60 days.
Fifty patients were treated (26 with GM, 16 with COL, 8 with both), and 102 patients were controls. Spontaneous eradication occurred in 7 percent of controls after 140 days median follow-up. Eradication rates were 42 percent with GM, 50 percent with COL, and 37.5 percent with combination (median treatment 31, 54, and 45 days, respectively). Eradication rates in the treatment arms were not statistically different, but all were significantly higher than the spontaneous eradication rate in controls. There were no observed side effects with either antibiotic.
CRE-positive clinical cultures and administration of intravenous antibiotics were not different between treated patients and controls. Overall mortality was higher in controls (53 percent vs. 22 percent; p<0.001), although CRE-related mortality was similar. Patients with successful eradication (spontaneous or after treatment) had lower mortality than those with persistent colonization (17 percent vs. 49 percent; p=0.002).
Oral GM and COL appear to be safe and effective in eradicating CRE gastrointestinal colonization when optimal infection control measures have failed, based on the study’s results. Although higher overall mortality was observed in controls, the authors did not adjust for differences in baseline characteristics (higher age, frequency of hematologic malignancies, and bedridden patients among controls). Therefore, gastrointestinal decolonization is not strongly correlated with better clinical outcome, and has to be balanced against development of resistance with prolonged therapy.
Is Amoxicillin Really Associated with Rash in Acute EBV Infection?
Reviewed by Christopher J. Graber, MD, MPH, FIDSA
The notion that amoxicillin, when given during an episode of acute Epstein-Barr virus (EBV) infection (infectious mononucleosis), leads to a rash in 80-100 percent of patients has been a widely believed clinical pearl since a relationship between ampicillin and rash in acute EBV was described in three studies published in 1967. However, two new studies recently challenge this assumption.
The first study
, published in Pediatrics
, was a retrospective examination of 238 children presenting with serologically confirmed acute EBV infection hospitalized at two tertiary medical centers in Israel from 1999 to 2009, 173 of whom were treated with antibiotics and 65 of whom were not. Fifty-seven (32.9 percent) of the antibiotic-treated patients developed a rash, as compared with 15 (23.1 percent) of the patients that did not receive antibiotics (p=0.16). Antibiotic-attributable rash, in which rash developed after administration of the first dose of antibiotic and up to 48 hours after the treatment ended, was seen in only 41 (23.7 percent) of the 173 patients who received antibiotics, though amoxicillin was associated with the highest rate of overall (39.3 percent) and antibiotic-attributable (29.5 percent) rash.
The second study, described in a letter to the editor
published in Clinical Infectious Diseases
, retrospectively identified 184 patients (either inpatient or outpatient) with acute EBV infection from a regional hospital in France from 2008 to 2013, 34 of whom had rash and 150 of whom did not. Fourteen (41 percent) of the patients with rash received amoxicillin, as compared with 71 (47 percent) of those who did not have rash. Furthermore, 16 (47 percent) of the patients with rash received any penicillin derivative, as compared with 87 (58 percent) of those who did not have rash.
The first study was limited because it only included hospitalized patients, and both studies were limited by relatively small sample sizes. However, they nevertheless suggest that the relationship between amoxicillin and rash in the setting of acute EBV infection may not be as well-defined as previously thought and that receipt of amoxicillin may not necessarily alter the likelihood of patients with acute EBV infection developing a rash. The authors of the first paper suggest that chemical differences between ampicillin and amoxicillin may be responsible for the differences between the studies from the 1960s and now, but the 1960s papers were also limited by relatively small sample sizes.
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:
- Updated European Guideline for Management of Clostridium difficile Infection
- Antibiotics for Low Back Pain?
- Occult Rifampin Resistance in Mycobacterium tuberculosis
- Autoimmune N-methyl-D-aspartate Receptor Encephalitis or Herpes Simplex Encephalitis?
- What Is the Linezolid Concentration?
- Administration of Antibiotic Therapy for 2 Days to Treat Possible Sepsis of Unknown Origin: A Pilot Study
< Previous Article |