By Carol A. Kemper, MD, FACP

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

How Dirty Is That Sandbox?

SOURCE: Orden C, Neila C, Blanco JL, et al. Recreational sandboxes for children and dogs can be a source of epidemic ribotypes of Clostridium difficile. Zoonoses Public Health 2017;1-8; doi: 10.1111/zph.12374.

You can blast public pools with chlorine, but it’s not really possible to do the same to the sand in public playgrounds. Fecal contamination from people, domestic dogs, and even wild animals, who may prefer the soft sand of a playground as a large outdoor toilet is a potential threat. Efforts have been made to figure out better ways to screen public sandboxes for parasites and other markers of bacterial contamination. For example, Toxocara canis, excreted by dogs, can have devastating effects in young children who ingest infected dirt. In May 2014, the CDC included T. canis on the list of “five neglected parasite infections in the United States” a classic sandbox infection.

Add Clostridium difficile to the list of possible sandbox infections. Various studies have found rates of C. difficile contamination of soil sampled from elementary school playgrounds, public parks, gardens, and cultivated fields ranging from 4% to 21%.

This study found a much wider distribution of C. difficile in Madrid playgrounds, with a diversity of strain types, affecting more than half of the child and dog sandboxes. The researchers examined 20 “pairs” of sandboxes designated for children and dogs, which were spaced 20 to 60 meters apart from one another, throughout the city parks. Samples of 200 grams of soil were collected at various points from each sandbox, and filtered specimens were cultured for C. difficile. Ribotype analysis and amplified fragment length polymorphism (AFLP) were used to distinguish isolates, and susceptibility studies were performed, including that for metronidazole heterogeneity.

More than half (52.5%) of the 40 sandboxes yielded C. difficile organisms, eight of which were toxigenic. Two of the toxigenic strains were isolated from children’s sandboxes and six from the dog sandboxes. While some of the strains were the same ribotype (mostly ribotypes 014 and 009, which are common in Europe), AFLP analysis confirmed that none of the strains were related. Five of the paired dog-child sandboxes in the same playground were contaminated, but only once with the same ribotype, and all with different strains. All of the isolates were broadly resistant to various antibiotics, and all were resistant to both meropenem and levofloxacin. None of the isolates exhibited heteroresistance to metronidazole.

C. difficile is no longer just a nosocomial pathogen — and, increasingly, patients coming into the hospital are a source for C. difficile and drug-resistant organisms. Our own community hospital in Mountain View, CA, screens all higher risk admissions for stool colonization with toxigenic C. difficile. In spring 2016, 73 of 455 higher risk admissions (16%) were PCR positive, and 19% of skilled nursing facility admissions were positive. (Higher risk admissions are defined as dialysis patients, patients admitted from long-term care or other facilities, and persons with a history of C. difficile.) Increasingly, the environment and food products are faulted for being sources for C. difficile colonization — but the source just as likely could be grandma or the household dog.

Rebuilding the Pyramids

As the number of new hepatitis C infections in the United States continues to escalate, in part fueled by the opioid epidemic, a consensus committee has called for major changes in the screening and treatment of hepatitis C infection in the United States.1,2 Even rural areas previously unaffected by hepatitis C virus (HCV) infection have seen spikes in the number of cases. Despite newer and highly effective antiviral therapy, the number of women aged 15 to 44 years with HCV infection doubled from 2006 to 2014, and an estimated 2.1 million women of reproductive age are now HCV-infected.3 While vertical transmission of HCV is not as likely as for hepatitis B virus, approximately 5-6% of women may transmit HCV infection to their infants.

Elimination of HCV infection has been proposed as a goal for the United States. With implementation of the following recommendations, the incidence of HCV infection within the United States could be reduced by 90%, and 90,000 hepatitis C-related deaths could be averted by 2030. These recommendations include:

  • Aggressive screening and treatment, especially of individuals who inject drugs, with a goal of treating 260,000 patients per year, without restrictions based on fibrosis score;
  • Improved access to needle exchange programs, especially in smaller communities and rural areas lacking these resources and expertise;
  • Improved access to opioid agonist therapy and treatment centers, especially in suburban and rural areas lacking these resources and expertise;
  • More aggressive treatment of identified cases. Only 5% of those without insurance or Medicaid receive treatment, compared with 46% of those with Medicaid and private insurance. Newer HCV treatment regimens are highly successful in injection drug users, although re-infection remains a problem. Reductions in the “community burden” of HCV infection will ultimately reduce the rate of new infections, and the cost of treatment provides substantial savings compared to the “public” cost of liver disease, cancer, and death.
  • Aggressive programs for those incarcerated with HCV infection. Prisoners are disproportionately affected by HCV. Data show that only 1% of prisoners with HCV infection receive treatment.
  • Programs to reduce the cost of HCV treatment, such as licensing for generic drugs for certain sectors of the population (such as Medicaid and prisons).
  • If > 260,000 persons per year could receive treatment, the above goal would be met before 2030.

Such programs can be successful, with Egypt being a prime example of what can be accomplished by a national campaign to reduce rates of active hepatitis C infection. In response to a nationwide epidemic of HCV infection, Egypt developed a National Strategy for Control for Viral Hepatitis in 2008, with a series of steps aimed at reducing cases of HCV infection by 300,000 per year, with a goal of reducing HCV infection to less than 2% of the population by 2025.

Licensing of technologies for the manufacture of antiviral medications has been a key part of this strategy, allowing developing countries access to certain medications for a fraction of the cost in the United States. (For example, sofosbuvir costs about $10 per day for a three-month course in Egypt, while the same medication costs about $1,000 per day in the United States.) To improve compliance with treatment and to reduce the risk of black-market sales, drugs are dispensed through pharmacies, through a tightly controlled system, where Egyptians can receive directly observed therapy, sometimes for free.

Note that the annual treatment goal of the consensus committee for the entire United States is less than that of the Egyptian government. No one is proposing to rebuild the pyramids, but can we do at least as well as the Egyptians with our HCV program?


  1. Buckley GJ, Strom BL. A national strategy for the elimination of viral hepatitis emphasizes prevention, screening, and universal treatment of hepatitis C. Ann Intern Med 2017;166:895-896.
  2. Talal AH, Thomas DL, Reynolds JL, Khalsa JH. Toward optimal control of hepatitis C virus infection in persons with substance use disorders. Ann Intern Med 2017;166:897-898.
  3. Ly KN, Jiles RB, Teshale EH, et al. Hepatitis C virus infection among reproductive-aged women and children in the United States, 2006 to 2014. Ann Intern Med 2017;166:775-782.

Resistant TB in India: Unrecognized Mutations

SOURCE: Manson AL, Abeel T, Galagan JE, et al. Mycobacterium tuberculosis whole genome sequences from Southern India suggest novel resistance mechanisms and the need for region-specific diagnostics. Clin Infect Dis 2017;64:1494-1501.

We increasingly depend on newer technologies to tell us when we have a case of possible drug-resistant tuberculosis. For instance, the Xpert system detects rifampin resistance by targeting select mutations within the RNA polymerase rpoB gene, effectively serving as a marker of multi-drug resistance (i.e., combined isoniazid and rifampin resistance). As additional mutations are recognized, this system may be expanded, but the question remains: What proportion of mutations are as yet unrecognized?

In India, 2-3% of new infections and 12-17% of re-infections are multi-drug resistant. Rates of isoniazid resistance are even higher, ranging from 10-15% of new infections and 30-40% of re-infections. An additional 4% of infections are XDR. In an effort to better understand the genetic diversity and the evolution of genetic markers of drug resistance in India, these authors performed whole genome sequencing of 223 randomly selected strains from 196 patients in southern India. The majority of the strains (70%) were lineage 1 (EIA or Indo-Oceanic lineage) and 16% were lineage 3 (LIN-3 or Central Asian lineage). These strains are significantly less common in other parts of the world.

After removing similar strains from the same individual, 201 strains were tested for phenotypic susceptibility. Resistant rates in descending order included: isoniazid (18%), streptomycin (9%), rifampin (6%), and ethambutol (3%).

A phylogenetic tree was constructed, placing strains within a previous set of 243 globally diverse MTb isolates. Although the library contained several lineage 1 and lineage 3 strains, they were substantially distant (> 500 years) from the study strains. The tree revealed “deeply branching clusters composed solely of southern Indian strains,” with few to no SNP differences. Six of the 22 clonal groups from this study contained strains from more than one individual, which were highly conserved, indicating person-to-person transmission. The authors commented that these individuals often were treated within the same facility, although it was not clear whether transmission occurred at the treatment facility or within the community. Resistance to INH arose independently for 33 isolates. Four of these isolates also were rifampin resistant. There was sufficient depth to the material to determine that INH resistance pre-dated rifampin resistance in all four cases.

Most interestingly, the authors determined how well two published lists of recognized mutations predicted resistance for the study isolates. The sensitivity of the two sets of published mutations for predicting INH resistance was fairly good at 74% and 73%, although lower than that described for isolates collected from the United Kingdom. The specificity was > 94%. However, sensitivities for the other drugs were more variable, ranging from 27-60% for streptomycin, 50-80% for ethambutol, and 73% for rifampin. Clearly, novel mechanisms or mutations drive some of the resistance in these southern Indian isolates.

The authors attempted to search for novel mutations associated with unexplained resistance, identifying several candidates. However, they were unable to identify any genotypic changes for three patients with serial isolates demonstrating progressive phenotypic drug resistance. They also examined strains from mixed infections, identifying seven strains with ambiguous calls at sites found within previously recognized resistance genes. They also found genetic ambiguity within genes associated with resistance that did not appear to explain additional phenotypic resistance.

Overall, at least one-quarter of phenotypic resistance could not be explained by currently recognized drug-resistant mutations. The authors caution that catalogs of drug-resistant mutations built on lineages or strains common in certain countries may not readily apply to strains from other parts of the world.