February 2019: General ID

Even in a year when the flu vaccine is “bad,” it still does a lot of good.  The 2017-2018 influenza season was a severe one, and the authors sought to quantify the effects of the seasonal influenza vaccine that year.  Investigators used age-specific national estimates of 2017-2018 vaccine coverage and disease burden data from a surveillance cohort of several thousand individuals to estimate the vaccine’s efficacy (VE).  Afterward, the authors extrapolated from those numbers to estimate the total reductions in influenza disease burden attributable to vaccination.

The authors calculate the 2017-2018 seasonal influenza vaccine to have had a VE of 38% against outpatient, laboratory-confirmed influenza disease.  In a year with 48 million episodes of influenza, this vaccine nonetheless prevented an estimated 7 million episodes of illness, 3.7 million doctor visits, 109,000 hospitalizations, and 8,000 deaths in the US (that’s still more than twice the number of Americans who have been killed by terrorism since the nation’s founding, which is weird because in 2018 the Department of Homeland Security’s budget was more than six times the budget of the CDC – but I digress).  Vaccine efficacy was highest in children under the age of four, among whom it prevented 2 million episodes of illness, 41% of expected hospitalizations, and 74 deaths.  The number needed to vaccinate to prevent one hospitalization was 1223 across all age groups, but only 821 for children under the age of four and 462 for adults over the age of sixty-five.

There’s something interesting here – if you look at the whisker plots of vaccine effectiveness by age group in figure 3, and specifically at the plots for the predominant H3N2 strain, the VE is way higher for kids under 4 than another other age group.  It’s not close; the 95% confidence interval barely overlaps with those of the other age groups.  So riddle me this, folks who actually stayed awake through their immunology classes: why would young children with developing immune systems mount better immune responses to a vaccine than older children and adults?  Is this evidence of the “immunologic fatigue” hypothesis wherein serial annual influenza vaccinations allegedly provide diminishing returns? 30715278


Now we have a clinical (and not just theoretical) reason to prefer the high-dose influenza vaccine in adults over 65: it more effectively prevents hospitalizations.  For years we’ve had a high-dose seasonal influenza vaccine that was approved but not preferentially recommended for adults over the age of 65.  In vitro data showed this vaccine produced better antibody titers in such patients, but we lacked data to suggest that those higher titers translated into any meaningful clinical benefit.  That’s changed in past few years. and so while this is not the first paper to show such a benefit, it does so compellingly.

The authors used the Veteran’s Health Administration database and Medicare files to compare the relative vaccine effectiveness of the standard-dose and high-dose influenza vaccines in veterans over age 65 across five influenza seasons (2010 through 2015).  In total, they included 3.6 million person-years of observation from 1.7 million veterans, 4% of whom received the high-dose vaccine and 96% of whom received the standard-dose vaccine.  They found that the adjusted relative vaccine effectiveness of the high-dose versus standard dose vaccine was 10% for all-cause hospitalization, 14% versus influenza-associated hospitalization, and 18% versus cardio-respiratory hospitalization, with no differences between influenza seasons.  As a control, the authors compared rates of UTI-associated hospitalization, and found – as expected – no difference between the high and standard-dose vaccinees.

Long story short: we have data that the high-dose seasonal influenza vaccine provides clinically important advantages over the standard-dose vaccine for older adults.  It’s time to stop recommending the “either-or” approach and to start actively preferring the high-dose flu jab for our older adults. 30745146


Humans are asymptomatic carriers of malaria species that infect monkeys.  The authors examined blood samples collected from 5422 people across 23 villages in western Cambodia as part of a malaria surveillance study. Using quantitative PCR, they found asymptomatic malaria parasite infections in 9% of the samples.  Of these, the authors discovered that 21 individuals living near forested areas had infections with species of malaria that typically infect long-tailed macaques, specifically P.cynomolgi and P.knowlesi (the latter of which also causes diseases in people). The mean parasite densities were 52,488 organisms/ml for P.knowlesi but only 3604 organisms/ml for P.cynomolgi , suggesting that P.cynomolgi may be less well suited for human hosts. 

Not only is this cool, but I wonder if Plasmodium cynomolgi might represent a novel approach for developing a human malaria vaccine.  We might be hard pressed to create an attenuated strain of P. falciparum, but what about inoculating people with a related species that just doesn’t replicate well in humans? 30295822

Most positive lyme IgMs represent false positives, and most of these result in inappropriate antibiotic prescribtions.  Investigators obtained all Lyme serologies ordered at US Air Force healthcare facilities over a four-year period.  All patients who had Lyme serologies were identified, and researchers collected available data on the patients’ documented exposure histories, clinical presentations, and followup Lyme serologies.  Based on standard criteria for adjudicating a positive lyme IgM, a test was considered a false positive if either the patient had no exposure risk (i.e. test ordered outside of tick season or in a state where Ixodes ticks are not endemic), the patient had only nonspecific or atypical symptoms of Lyme disease, or a followup test within 30 days of the initial IgM was negative.

Over 18,000 Lyme serologies were sent on nearly 16,000 patients during the study period.  A total 819 (4.8%) of the initial tests were positive or equivocal, of which 92% were appropriately reflexed to immunoblot; in addition, another 599 immunoblots were ordered without initial testing or despite an initial test procuring a negative result; in total, 25% of Lyme serologies were ordered inappropriately.  Of all the immunoblots performed, 23% were positive, and of the positive IgM immunoblots, 53% proved to be false positives.  Adults (58% vs 36%; p=0.008) and women (66% vs 42%; p<0.001) were more likely to have false positive tests.  Despite being false positives, 81% of patients in this category received antibiotics anyway.

Conclusion?  No one knows how to test for Lyme disease appropriately, which means lots of people get treated inappropriately.  I thank my lucky stars I live in an area with no endemic Lyme disease, and so usually don’t have to think about this headache of a diagnosis. 30802651