By Carol A. Kemper, MD, FACP

Clinical Associate Professor of Medicine, Stanford University, Division of Infectious Diseases, Santa Clara Valley Medical Center

Dr. Kemper reports no financial relationships relevant to this field of study.

Metagenomics for Prosthetic Joint Infection

SOURCE: Thoendel MJ, Jeraldo PR, Greenwood-Quaintance KE, et al. Identification of prosthetic joint infection pathogens using a shotgun metagenomics approach. Clin Infect Dis 2018;67:1333-1338.

A good proportion of prosthetic joint infections (PJI) are culture-negative, necessitating prolonged courses of empiric antibacterial therapy. Even more frustrating are those cases of suspected PJI in which infection cannot be confirmed, and the role of infection in chronic hip or knee pain or aseptic loosening remains a concern.

These authors compared culture data derived from sonication of resected hip and knee components with a shotgun metagenomics approach. In this case, “shotgun” simply means that all nucleic acid in a sample is extracted and sequenced by next-generation sequencing techniques, and the sequences are read and identified.

From 2011-2016, a total of 408 cases were selected sequentially from a database of orthopedic cases in which sonication for culture was used, although the decision to use sonication for any particular case was at the discretion of the individual surgeon. These cases included 213 patients with PJI and 195 without infection in whom surgery was performed for other reasons (e.g., aseptic loosening). A positive culture was defined as isolation of the same organism from two or more intra-operative specimens, with at least 20 colony-forming units of growth.

Of the 213 subjects with suspected PJI, 115 (54%) were culture-positive by sonication and 98 (46%) were culture-negative. Metagenomics identified the same bacterial pathogen as sonication cultures in 109 of 115 (94.8%) cases. Additionally, metagenomics identified probable pathogens in 11 of these culture-positive cases (9.6%) not previously detected by culture, suggesting polymicrobial infection. Potential pathogens also were identified in 43 of 98 (43.9%) culture-negative cases of PJI. While most of the organisms identified by metagenomics were organisms common to PJI, one unique pathogen picked up in metagenomic testing, not identified by culture, was Mycoplasma salivarium.

However, when comparing all available culture results, including the results of blood culture or any cultures obtained prior to surgery, metagenomics identified pathogens in 121 of 146 culture-positive cases (82.9%), plus additional pathogens detected in 12 cases. Sixteen cases had at least one pathogen isolated in culture that was not detected in joint specimens by metagenomics; 14 of these had received antibiotics prior to surgery, and seven also had negative sonification cultures.

In only six of the 115 PJI cases (5.2%), an organism was detected by sonication culture but was not detected by metagenomics. These included the isolation of Pseudomonas aeruginosa infection in two cases (which may be more susceptible to DNA degradation), Candida albicans in two cases, and Escherichia coli, Enterobacter cloacae, and Mycobacterium abscessus in one case each.

Sonicated specimens from (presumably) uninfected prosthesis or components were submitted for metagenomic testing in 195 cases. Seven (3.6%) were positive for potential pathogens, all organisms recognized to cause PJI.

The authors acknowledged that a limitation to the use of metagenomics for PJI is the ability to distinguish background “noise” and contaminants from true pathogens. Many of the organisms contributing to PJI generally would be considered contaminants in other clinical specimens (e.g., skin organisms). Higher “read counts” provided an inference as to the likelihood of a true pathogen, but generally those specimens that were negative in culture also gave fewer “read counts.” Further, although metagenomics may provide information about the presence of a potential pathogen, susceptibility data still are lacking.

Helicobacter pylori: A Mini Primer

SOURCE: Siddique O, Ovalle A, Siddique AS, Moss SF. Helicobacter pylori infection: An update for the internist in the age of increasing global antibiotic resistance. Am J Med 2018;131:473-479.

Like every other infection we deal with, Helicobacter pylori (HP) is increasingly drug resistant. Estimated failure rates are 5-10%, even after receipt of two different antimicrobial regimens. Failures most often are due to resistance to clarithromycin (which may be as high as 30% in some countries and in some parts of the United States) and levofloxacin (which also may be approaching resistance rates of 30% in some parts of the United States). Physicians need to keep pace with the consequences of this development and newer recommendations. Although the prevalence of HP seems to be decreasing in the United States, at least in higher socioeconomic strata, HP remains a problem for lower-income groups, travelers to developing countries, and the rest of the world. The prevalence of HP is believed to be > 50% in some parts of the world, especially in Central Asia, Central America, and Eastern Europe.

There are multiple barriers to appropriate testing and treatment:

  • The first barrier is the promotion of testing for HP in patients at risk. HP screening is indicated for anyone with recurring epigastric discomfort, chronic use of nonsteroidal anti-inflammatory drugs, unexplained iron deficiency anemia, and ITP. Any of the “alarm symptoms,” such as recurrent vomiting, weight loss, and dysphagia, especially with a family history of gastric cancer, should prompt endoscopy with biopsy and examination for HP.
  • The second barrier is the type and timing of testing. Tests for active infection include stool antigen testing and the urea breath test (both 95% sensitive, 95% specific). But they must be performed more than four weeks after the use of any bismuth-containing product or antibiotic, and all proton pump inhibitors (PPI) must be stopped for more than two weeks. Serologic testing is not recommended, as it may remain positive for many years after successful eradication and is associated with a higher false-positive rate.
  • When selecting a first-line regimen, the patient should be queried about antibiotic use in the past one to two years. Prior treatment with macrolides or levofloxacin may increase the risk of resistance, and a regimen without the respective agent should be selected. In contrast, resistance to amoxicillin, tetracycline, and metronidazole is uncommon (< 2%).
  • Patients should be counseled that strict adherence to the regimen is necessary. Missed doses may increase the risk of developing resistance during treatment, especially with clarithromycin; and completion of the entire regimen is important to successful eradication.
  • In those who fail a first-line regimen, consider whether the patient has a penicillin allergy and whether clarithromycin or levofloxacin was used in the first regimen. There are two or three options depending on the answers to these questions. For example, for patients who fail a first-line regimen containing clarithromycin, a regimen without clarithromycin should be selected (e.g., amoxicillin, levofloxacin, PPI × 14 days).
  • The management of patients who have failed two regimens is not straightforward. Endoscopy with biopsy and culture for susceptibility testing is recommended, although the organism does not always grow well in culture, and susceptibility testing is not widely available. Empiric treatment with a non-clarithromycin-based regimen (e.g., rifabutin, amoxicillin, PPI × 10 days) in those previously treated with clarithromycin can be attempted. A levofloxacin-containing regimen can be used in those not previously treated with fluoroquinolones. Increasing the dose of the PPI, or using newer, more potent PPIs, may be helpful.
  • Finally, confirming eradication four weeks or more following completion of treatment is mandatory.