By Carol A. Kemper, MD

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

Antibiotic Increases Honeybee Mortality

SOURCE: Raymann K, Shaffer Z, Moran NA. Antibiotic exposure perturbs the gut microbiota and elevates mortality in honeybees. PLoS Biol 2017;15:e2001861.

The investigators examined the gut microbiome of wild honeybees fed tetracycline, and determined the mortality of treated vs. untreated bees. Honeybee gut flora is highly conserved, meaning honeybees throughout the world share similar bacteria, probably as a social phenomenon. It has been theorized that this gut flora is essential to honeybee health and natural resistance to opportunistic pathogens. Metagenomic analysis of colonies with colony collapse disorder (CCD) have shown an increase in certain gut bacteria while other “core” bacteria are diminished (e.g., Bifidobacteria, Alphaproteobacteria, Lactobacillus) when compared with healthy hives.

The authors collected healthy honeybee hives, brought them into the laboratory, and fed them sterile sucrose syrup with or without tetracycline for five days before placing the colony back in the hive. Bees were sampled at different time-points post-antibiotic exposure. Tetracycline treatment resulted in significant changes to gut flora, with reductions in the relative and absolute abundance of eight core bacterial species. None of the core bacteria were completely eliminated, but Lactobacillus species, FIRM-4, FIRM-5, and Bifidobacteria were substantially reduced, as well as a gram-negative bacilli called Snodgrassella alvi. In contrast, a relative increase in Gilliamella apicola was observed in treated bees. Even at seven days post-treatment, significant reductions in normal gut flora persisted. Keep in mind that worker honeybees (who are all female) only live about four to six weeks, and only the largest and oldest bees are sturdy enough to carry their burden home at the end of the day.

Once honeybees were transplanted back to their hives, recovery of foraging bees was only 32% in treated hives compared with 64% in untreated hives meaning only half the number of the treated bees returned home at the end of their day compared to untreated controls. More mature bees also died in the treated hives compared with untreated hives.

The authors exposed treated bees to a strain of Serratia, a potential opportunistic pathogen, which appeared to reduce survival further in both age-controlled and non-age-controlled Kaplan-Meier analysis.

The basis for colony collapse disorder is not yet known. Preliminary work suggests it is the result of a complex interplay of multiple factors. Apiarists in some countries have tried various anti-mite treatment, as well as antibiotics, hoping to keep larval bees alive. Unfortunately, tetracycline may have an adverse effect on honeybee gut flora and survival. From personal experience — the effects of CCD are devastating — I’ve kept up to seven hives at a time, and overnight, the bees seem to vanish. I have given up on beekeeping, leaving it to those more expert than myself, hoping they will have better success. For one, it’s become quite expensive: A five-pound package of bees used to cost $15, mailed by USPO, and a queen in a matchbox was $2. Now, a starter hive costs upwards of $150. The Obama administration pushed for further research in honeybee mortality. Let’s hope the Trump administration continues this funding.

Sustained Hepegivirus Infection in Injection Drug Users

SOURCE: Kandathil AJ, Breitwieser FP, Sachithanandham J, et al. Presence of human hepegivirus-1 in a cohort of people who inject drugs. Ann Intern Med 2017; doi: 10.7326/M17-0085. [Epub ahead of print].

Molecular “deep diving” using such techniques as next-generation metagenomic sequencing (NGMS) is resulting in the discovery of new viruses almost every year. Previously, Kapoor and colleagues identified a novel RNA flavivirus in two individuals with hemophilia who had received multiple blood products.1 Viremia persisted in one of these individuals for 5.4 years without apparent symptoms or disease. This new virus was characterized as human hepegivirus, or HHpgV-1, which shares features with hepatitis C virus and the apparently non-pathogenic human pegivirus, HPgV, formerly called hepatitis G virus or GBV-C. HHpgV-1 also was found in two individuals who had received blood transfusion, but both individuals had cleared their viremia in subsequent specimens. Researchers now are sequencing and comparing these viruses with other similar viruses from chimpanzees, rodent variants, and several different groups of pegiviruses and hepaciviruses from bats to form a phylogenetic tree.

Armed with this information, Kandathil and colleagues examined serum samples obtained from a prospective national clinical cohort study for persons using injection drugs who received interferon for HIV/HCV co-infection. To enhance the likelihood of interesting findings, samples were selected from those previously known to be positive for GBV-C RNA (n = 20), SEN virus DNA (n = 24), HCV RNA (n = 42), and HCV antibody but not HCV RNA (n = 20). Plasma samples were obtained before and after interferon administration at 72 and 168 hours and stored.

Using these samples, NGMS generated 600 million reads, most of which were the expected HIV and HCV sequences. But a small number of samples generated different reads leading to the discovery of HHpgV-1 in four samples from two persons. These reads were detected both pre- and post-interferon, and the reads from one participant were so numerous as to be nearly sufficient to assemble a complete genome.

This led to the creation of a specific quantitative PCR, which was used on a subset of 156 individuals with a history of injection drug use. Seventeen (10.9%) were found to have HHpgV-1 in plasma samples. HHpgV-1 viremia was detected in eight of these individuals at two different time-points spanning a median of 5,886 days, suggesting that these individuals were viremic, on average, for many years (or repeatedly re-infected). Further testing of another six individuals at three different time-points confirmed sustained viremia for a median of 4,538 days (range, 1,524 to 6,158 days).

The implications of this finding and how long HHpgV-1 may have been in circulation in human beings are not known. Since more than half of the participants in the original cohort study had sustained HHpgV-1 viremia and active HCV infection, it is not known whether the viruses simply coexist or are, in some way, interdependent — or whether their presence is simply a marker of ongoing risky behavior. What blood banks are going to do with all this new information is even less clear.


  1. Kapoor A, Kumar A, Simmonds P, et al. Virome analysis of transfusion recipients reveals a novel human virus that shares genomic features with Hepaciviruses and Pegiviruses. MBio 2015;6:e01466-15. doi: 10.1128/mBio.01466-15.

Decrease SSIs: Take August Off

SOURCE: Anthony CA, Peterson RA, Polgreen LA, et al. The seasonal variability in surgical site infections and the association with warmer weather: A population-based investigation. Infect Control Hosp Epidemiol 2017; doi: 10/1017/ice/2017.84. [Epub ahead of print].

Previous studies have hinted at an increase in surgical site infections (SSIs) in the summer months. To determine the seasonal variation of SSIs, the investigators examined a large population database (from the Healthcare Cost and Utilization Project by the Agency for Healthcare Research and Quality) for all hospital discharges with a primary diagnosis of SSI from 1998-2011. This database contains data from 20% of nonfederal acute care hospitals in the United States (but does not contain outpatient data). Ten surgical procedures were the focus of the investigation, including colon resection, small bowel resection, knee arthroplasty, hip replacement, lower extremity fracture repair, spine fusion, cesarean section, inguinal or femoral hernia repair, another hernia repair, and exploratory laparotomy. Cases were aggregated by month, year, region, sex, age, and type of institution. Monthly average ambient temperatures were determined for each hospital location. Subgroup analysis with logistic regression was performed using two separate models.

After eliminating cases without adequate data, the sample size was 55,665,828 procedures performed at 2,512 unique hospitals. The monthly incidence of SSI clearly peaked in August of every year, and nadired in January, with an average difference of 26.5% (P < 0.001). SSI cases had higher rates of diabetes (19% vs. 14.7%) and obesity (9.5% vs. 5.6%) compared with controls, but were otherwise similar with regard to mean age, comorbidities, latitude, and region. Further, the odds of SSI were similar whether the procedure was performed in a teaching hospital or a nonteaching hospital.

After controlling for demographics and hospital co-factors, the odds of SSI increased by 2.1% for every five degrees Fahrenheit increase in average monthly temperature.

Remarkably, the increased odds of SSI from the warmest month to the coldest month (January) was double the effect of diabetes. And if July and August cases were eliminated from the database, the rate of SSI would decrease by 20.6%.

As yet, there is no good explanation for this observation. Is it really because of changes in the weather? Do people sweat more in the hospital? Are bandages not as secure in the heat? What I did not understand was the study focused on in-hospital SSI data, not outpatient data. Presumably, most hospital temperatures are controlled, with air conditioning in the summer months and heating in the winter. Shouldn’t the “ambient” temperature within hospitals be similar, more or less, regardless of whether you are in Minneapolis in January or Tucson in August? Some have theorized that the introduction of new house staff in July-August might affect SSI rates, and yet there was no apparent difference between teaching and nonteaching facilities in this study. Is there simply reduced or less capable staffing in hospitals in July-August because of vacations? Staff dreaming of that fish they caught last weekend and not focusing on dressing changes? Whatever the reason, this study suggests that if we were more like the French, took August entirely off, and moved all elective surgical cases to the winter months, the total annual burden of SSI would be reduced by approximately 20%.