Friday, December 18, 2009

Malaria: Forcing us to destroy our own brains...

From a recent PLOSOne study, some interesting findings on malaria pathogenesis. What we know is that getting cerebral malaria is both very bad and very unpredictable, so that it's very difficult to decide which patient will require closer monitoring than others. Management is non-specific and supportive, and we still don't exactly know why it happens. There are a lot of theories out there, many of which center around the sludging of blood in the cerebral vessels, causing decreased brain blood flow and the symptoms we see. This has always been suspect, since there are a great deal of inconsistencies with this hypothesis, so investigation has continued.

Adding to some prior work that they've done, these investigators looked at a population in India where malaria is endemic. From screening the serum of patients with both severe malaria and cerebral malaria, they found some interesting differences. Notably, the patients with cerebral malaria had a specific cytokine response that seemed to induce a reaction to a series of brain-specific proteins in the form of antibody production. What the effect of this auto-antibody production is on the symptoms seen in cerebral malaria remains unclear, especially since their concentration wasn't related to disease severity. Determining whether these antibodies contribute to disease, or serve as simply a marker for disease progression or existence, still has to be sussed out. Interestingly, however, the strongest signal was to a different brain protein compared with their prior study in African patients, suggesting a variant host response, although there was some overlap in this data.

Getting more people into this study would have powered their results a bit more, and perhaps helped pinpoint the brain protein a bit better, but nevertheless, it's an interesting theory that deserves some more study, since it has sizable implications, both for therapeutic and for prognostic purposes. Given that we are making only slow progress in the battle against this disease, the more information and research on the topic, the better.

Bansal, D., Herbert, F., Lim, P., Deshpande, P., Bécavin, C., Guiyedi, V., de Maria, I., Rousselle, J., Namane, A., Jain, R., Cazenave, P., Mishra, G., Ferlini, C., Fesel, C., Benecke, A., & Pied, S. (2009). IgG Autoantibody to Brain Beta Tubulin III Associated with Cytokine Cluster-II Discriminate Cerebral Malaria in Central India PLoS ONE, 4 (12) DOI: 10.1371/journal.pone.0008245

Saturday, December 5, 2009

Justifying my existence

As somebody training in both critical care and infectious diseases, I'm often left trying to explain what I hope to do with my career, struggling with coherent answers to both laypeople and other doctors. People get sick because of infections. Also, sick people end up getting infections. That's how I look at it, and hence my training.

This study, published by some of my bosses, helps validate all of this. Looking at one day in ICUs across the planet, they captured a point-in-time, just to see what the burden of 'infection' is in our planet's ICUs. Obviously, there are huge problems with potential reporting bias and the like, but it's as good a global survey as you can justifiably get. And the results are interesting, showing that half of all patients in ICUs were considered infected, with these patients doing significantly worse overall. Nearly three quarters of the patients were receiving antibiotics, some just for propylaxis. Also, the longer that you were sick, the more likely you were to have an infection. There was also a correlation between national GDP and rates of ICU infection, which, once again, isn't surprising, since we know that infection control procedures aren't always cheap

The types of bacteria and infections reported were interesting, with pneumonias as the most common infection, and a reasonable distribution of types of bacteria. There is an increasing prevalence of dangerous gram-negative bacteria, and a persistently worrisome rate of antibiotic-resistant bacteria. I've talked about this quandary before, since as we're able to care for sicker and sicker patients, we're also using more and more antibiotics, leading to all sorts of resistance problems for future generations. Also, what constitutes an 'infection' is a difficult distinction, one that we typically err on the side of caution by treating. In the end, with very few new antibiotics coming down the research pipeline, we will face a tremendous challenge outwitting these ever-shifting, diverse group of organisms.

Vincent JL, Rello J, Marshall J, Silva E, Anzueto A, Martin CD, Moreno R, Lipman J, Gomersall C, Sakr Y, Reinhart K, & EPIC II Group of Investigators (2009). International study of the prevalence and outcomes of infection in intensive care units. JAMA : the journal of the American Medical Association, 302 (21), 2323-9 PMID: 19952319

Monday, November 23, 2009

Back to the flu-ture

It's been quite some time since I've been here. I've moved cities and changed jobs, so it's taken some time to get settled. Anyway, I'm hoping to be more active over the next little while.

This pesky influenza virus that people seem to be talking about all of the time has kept me busy as well. We're finally seeing some downswings over the past two weeks, both locally here in Toronto and across North America , and hopefully, after the expected second phase of H1N1 later on in the winter and more people obtain immunity, this pandemic will slowly die down. Unfortunately, our yearly visit from plain ol' seasonal influenza is still to come, and it will be very interesting to compare mortality rates from the two viruses, just as a marker of the effectiveness of our collective panic, especially since our treatment options for plain ol' influenza are very different than big bad pandemic influenza.

Since I'm tired of pandemic strain H1N1, I'll go back to our original fear-monger, H5N1. Remember him? He's the one that everybody was worried about for years, poking his head out in various parts of the Asian land-mass, killing a few people, then going back to spend his time with chickens and other birds. This paper (see ref below), explores that exact concept in a group of poultry farmers in Vietnam, the highest-risk group of all high-risk groups for infection. In 2005, at the height of H5N1 paranoia, Vietnamese public health officials sampled the blood of chickens and/or ducks in all the farms in an endemic region near Ho Chi Minh City. Any farm with any H5N1 had its fowl culled, and, after some time, the farmers were all tested for antibodies to H5n1. Easy enough.

They found, confirming findings from other parts of the world, that fowl-to-human transmission of H5N1 is very, very rare, in that only 3 of the 500 people tested had any evidence of immunity, and those only had very low-levels, which may represent cross-reactivity to another influenza virus. We know that humans aren't easily infected with this virus, in its current form, and it would take some fairly drastic antigenic shifting for us to be affected. Nothing new, or surprising, from this study, but just more evidence that the currently circulating H5N1 virus does not easily transmit to humans, even with significant and direct exposure to the virus.

So, wash your hands, get your flu shots, live your life, and don't let the media freak you out about pandemics any time soon.

Schultsz, C., Van Dung, N., Hai, L., Quang Ha, D., Peiris, J., Lim, W., Garcia, J., Dac Tho, N., Thi Hoang Lan, N., Huu Tho, H., Xuan Thao, P., van Doorn, H., Vinh Chau, N., Farrar, J., & de Jong, M. (2009). Prevalence of Antibodies against Avian Influenza A (H5N1) Virus among Cullers and Poultry Workers in Ho Chi Minh City, 2005 PLoS ONE, 4 (11) DOI: 10.1371/journal.pone.0007948

Thursday, May 21, 2009

Progress in global health

The biggest global health project in the history of of global health, the Millennium Development Goals, just released its first interim report on the health achievements. The results are mixed. While child mortality has come down nicely, there's still large strides to make in drug availability, adolescent pregnancy rates, newborn/maternal mortality and especially in the distribution of many of the interventions, since the poorest of the poor still seem to be left out of much of the advancements made. There's been a lot of focus upon maternal mortality, and to see no significant improvement in the 20 years (1990 is the baseline year) is disheartening.

More info is here:

Tuesday, May 19, 2009

Back to birds and flu

A PLOS One article, looking at how our current practice of influenza vaccination leaves us vulnerable to an H5N1 pandemic...

Let's put aside the (current) H1N1 concern, and go back to the H5N1, or avian, influenza strain, and its potential for a bad pandemic, since its lethality is significantly higher than any other currently circulating strain. We know that our current seasonal influenza vaccine does not protect us against infection with the H1N1 strain currently being passed around the globe. But not protecting us and leaving us more vulnerable are two very different things, something that flummoxes a lot of public health experts. What this paper did, roughly, was the following: immunize some mice against the H3N2 influenza, one of the components of every seasonal flu vaccine; document protection, both by antibody levels and by experimental infection with that H3N2 virus; infect with H5N1, and see which mice do best. All of the mice initially became ill, but the ones that were previously infected with H3N2, but not previously vaccinated, recovered at about day 6. Both the mice that were vaccinated against H3N2 and the mice that weren't previously experimentally infected with H3N2 didn't do so well, almost all having to be euthanized.

What conclusions can be drawn? Vaccination against one strain of influenza does not always protect us against another strain. This we knew. When naturally infected, our bodies mount a broader and more powerful immune response than when vaccinated against specific antigens, which we know very well. Vaccination, and subsequent protection against natural infection with seasonal influenza, leaves mice more vulnerable to a different strain, ie avian flu.

We shouldn't reconsider the principle of universal seasonal influenza vaccination - it's benefits are very well documented - just for the fear of an impending avian flu pandemic. What we should do, however, is investigate other vaccine subtypes and see if this can be replicated, ie does H1N1 (another component of our current vaccine) vaccination leave us more vulnerable? Can we alter, or expand, the subtypes in our seasonal vaccine so that we wouldn't be left more vulnerable to a novel strain? And most importantly, can we find a way to speed up vaccine design and dissemination, so that we are more prepared for a quickly spreading pandemic?

While the current swine flu seems to be more alarm than threat, it has gotten the global public health cylinders in motion, and has brought these discussions back to the forefront.

Bodewes, R., Kreijtz, J., Baas, C., Geelhoed-Mieras, M., de Mutsert, G., van Amerongen, G., van den Brand, J., Fouchier, R., Osterhaus, A., & Rimmelzwaan, G. (2009). Vaccination against Human Influenza A/H3N2 Virus Prevents the Induction of Heterosubtypic Immunity against Lethal Infection with Avian Influenza A/H5N1 Virus PLoS ONE, 4 (5) DOI: 10.1371/journal.pone.0005538

Saturday, April 25, 2009

Men with infected scrotums less desirable to women!

Stating the obvious, but still nice to have the data. Ours being a shallow society, the 'marriageability' of somebody with a filarial hydrocele (only click if you're not eating your breakfast and you have a strong stomach) is probably not that high. The severe impact on sexual function, as well as the obvious cosmetic challenges, make them low on the list of potential suitors for young ladies.

That's exactly what researchers in Orissa, India found, published in PLOS NTD this week, when they asked a community their thoughts on men with hydroceles, both sufferers and non-sufferers. Some background: filariasis is an mosquito-borne parasitic disease that is exceedingly common in tropical regions, India being especially affected. It classically causes elephantiasis, but can also cause hydrocele, a swelling of the scrotum. In endemic areas, most people are first infected in childhood, but don't manifest symptoms until later in life with a greater parasite load, typically between the ages of 18-30, the prime years for marriage. Unfortunately, once it's reached the point where it's visible and impacting one's life, there's likely enough scarring and lymphatic damage that medicines aren't helpful, and surgery is still unproven, even when available.

What the researchers in this study did is ask the community how they felt about people with hydroceles. The results are unsurprisingly sad. 94% of wives of patients were dissatisfied with their sexual life, and that these men are overwhelmingly the 'last choice' for marriage. 94% of the patients themselves reported sexual frustration, with 88% reporting severe pain during intercourse. The morbidity of this disease is clearly profound, and most of the sufferers don't have appropriate psycho-social support groups to help them out.

The best option we have, then, is for its prevention or early treatment. There's been a movement over the past few years towards eradication of the disease, with the Gates foundation being large funders. There are no other animal hosts, so if we treat it in humans, then we can conceivably control it altogether. Vector control through mosquito control is a challenge, as any malaria expert will tell you, although that is one focus, affecting both diseases. Our best hopes are through mass drug administration: in some places, once a year, in others, added to salt, with the hope of bringing down community-wide levels of parasite to slow down transmission and move towards eventual eradication. Significant progress is being made, so if you have a few extra dollars, make a donation - they're a great bunch of people, and are scaling up their projects in creative and exciting ways. I'm in India right now, working at a hospital, and have seen a few horrible cases of filariasis, and, as the researchers have shown, it's a debilitating disease.

One of the worrisome parts of the eradication campaign is that about 10% of filariasis is caused by the parasite B. malayi, while 90% is caused by W. bancrofti. The eradication campaign is effective against the latter, but not against the former, which has a large animal reservoir. Let's hope that we don't see a bump in filariasis incidence after we've eliminated W. bancrofti. It's unlikely, but something to be cognizant of as we move forward.

Babu, B., Mishra, S., & Nayak, A. (2009). Marriage, Sex, and Hydrocele: An Ethnographic Study on the Effect of Filarial Hydrocele on Conjugal Life and Marriageability from Orissa, India PLoS Neglected Tropical Diseases, 3 (4) DOI: 10.1371/journal.pntd.0000414

Friday, April 24, 2009

More on pigs and flu

The CDC is stepping up its surveillance for swine flu (see prior post), and five more cases have been reported from California and Texas, none with known swine exposures, and a couple with known human-human contact.

To follow this, go to the CDC's swine flu website here.

Centers for Disease Control (2009). Human Swine Influenza Investigation CDC

Thursday, April 23, 2009

Amoxicillin with your froot loops

I saw this Stop & Shop ad in Boston while doing my grocery shopping a few months ago. At first, I thought it was a joke: I mean, free antibiotics? After getting clarification from the pharmacist, the scheme is that if you bring in a prescription for an antibiotic, they will give you the generic version, for free. They'll decide on the course - up to 14 days for some antibiotics!

The problems with this, obviously, are tremendous. Not just being exposed to an antibiotic unnecessarily (and they all have side effects). Not just more vociferous demands on health care providers to give them an antibiotic prescription for their runny nose, with the knowledge that they'll be free. But mostly, and I haven't seen the data (and I doubt that I ever will), it's the likelihood that this will lead to an over-dispensation of antibiotics. Which is a very bad thing.

Antibiotic resistance is one of the gravest challenges facing our society over the next generation. They were the opening salvo of the modern medical revolution half-a-century ago, and are directly responsible for much of the healthcare gains that we've made as a society. Unfortunately, bacteria are smarter than we are, and coming up with new antibiotics proceeds at only a snail's pace.

While the new drug pipeline continues to flow, what's most important is that we try our hardest to keep our current armamentarium as active as possible, with judicious and intelligent usage, effective infection control of known resistant organisms, and improved santitation and hygiene in high-risk environments. Otherwise, it would be a scary world if we continue to see the spread of resistant bacteria without the drugs to treat them with.

The Lancet Infectious Diseases, . (2009). US supermarkets redefine antibiotic misuse The Lancet Infectious Diseases, 9 (5), 265-265 DOI: 10.1016/S1473-3099(09)70115-X

Wednesday, April 22, 2009

Of Pigs and Flu

There's been some press recently about the recent report by the CDC of two children recently infected with swine influenza. Two children, without any discernible link to either pigs or each other, independently presented within a week of each other in San Diego with flu-like illness. Both were tested for influenza, found to be unsubtypable (not one of the common subtypes that we test against routinely), and sent off to the CDC for further testing, where it was found to be swine influenza, H1N1. Both children got better with supportive care.

There's a number of reasons why this is notable:
-all documented cases of swine influenza in humans over the past few years in the US have come with a documented exposure to swine. If this represents human-to-human transmission of the virus, then it's a feature of this particular virus that would be novel
-this is a novel virus, in that it's combination of genes are unique amongst influenza viruses in various databases around the world. Another example of the mutability of the influenza virus, it's likely a combination of known avian, human, and swine viruses.
-they are resistant to one of the major classes of anti-virals, and testing for the other class is underway

The main implications are that we need to constantly be vigilant against the flu, not just the seasonal variety. Also, surveillance systems like through which we obtained the data about these two cases are crucial, in all parts of the world, to try and get a step ahead of these constantly adapting viruses.

Swine Influenza A (H1N1) Infection in Two Children --- Southern California, March--April 2009 CDC Dispatch.

Centers for Disease Control (2009).

Sunday, April 19, 2009

Changing how we treat tuberculosis

Tuberculosis is a tough disease. Once infected, you're committed to months of a hodge-podge of medications (yes, i used the word hodge-podge) with varying side effects. What we've settled on is a standard four-drug regimen, using four drugs mostly to prevent against the development of resistance, a phenomenon that I'll talk about in a later post. This standard course of therapy is riddled with compliance issues, making a shorter, more tolerable regimen in everybody's best interests. And finding new drugs has been incredibly difficult.

So it's refreshing to see this study from Brazil, where they randomized treatment-naive patients to 8 weeks of either ethambutol, the standard, or moxifloxacin, a fluoroquinolone antibiotic that we know works against TB, in addition to the classic three-drug regimen. They then finished the 6 months of therapy with the standard two-drug regimen. They excluded patients with HIV requiring antiretroviral therapy and patients whose initial culture grew out a multi-drug resistant organism, among other things. The findings are striking.

At 8 weeks of treatment, patients in the moxifloxacin group had an 80% culture negative rate, compared with 63% in the ethambutol group, with the difference noticeable even at the 1 week point. There were no significant differences for adverse events.

Intensive treatment of TB was initially 18 months, then 9 months, then 6 months, where it is now. If a shorter course of therapy with moxifloxacin is compared with a longer course and these results hold, then we may be looking at an even shorter course of therapy, a significant change. If moxifloxacin's effects are in just killing a bunch of TB at the very beginning and leaving the slow-growing bugs to persist, then you'd expect a difference in relapse rates, which there wasn't in the study, although it may not have been a big enough study to detect a difference in a relatively rare finding.

The major goal for TB therapy globally is simplification: with this, you'll achieve resistance control, be able to manage disease in the community and not only in the major centres, and get more people completing their treatment regimen. This may be a step towards that goal.

Conde, M., Efron, A., Loredo, C., De Souza, G., Graça, N., Cezar, M., Ram, M., Chaudhary, M., Bishai, W., & Kritski, A. (2009). Moxifloxacin versus ethambutol in the initial treatment of tuberculosis: a double-blind, randomised, controlled phase II trial The Lancet, 373 (9670), 1183-1189 DOI: 10.1016/S0140-6736(09)60333-0

Wednesday, April 15, 2009

All is infectious

One of my professors in medical school a few years back told me that, in the end, almost every disease will be proven to have some infectious trigger or component. We scoffed, being the know-it-all med students that we were, but the data is accumulating on his side.

The theories on coronary disease and chlamydia. Crohn's disease and mycobacteria. All of the cancer-causing viruses out there. No damning evidence yet for ALL diseases, but definitely something to consider in our microbe-laden world. Add diabetes to that list.

The first study I saw on this front was in JAMA last year, which looked at an interesting subtype of diabetes, mostly present in sub-saharan Africa. These patients would be especially prone to episodes of ketosis and severe hyperglycemia, with intervening periods where they wouldn't require insulin and have normal sugars. They found a fairly impressive rate of human herpes virus 8 infection at disease onset, compared with non-diabetic controls and controls with typical diabetes. It hasn't been replicated in other populations yet, but work on this is pending.

Adding to this are two studies discussed in an editorial in JAMA this week. The first found, from pancreas samples obtained at autopsy, much higher rates of enterovirus infection in the pancreases of type 1 diabetics compared with non-diabetics. The pancreatic protein expression profile was also consistent with a chronic viral infection. Further, the second study looked at rates of diabetes in a large sample of people, with those who have a variant in the gene IFIH1, a well-known enzyme that's involved, among other things, in our response to picornaviruses, of which enterovirus is one. Those who had a rare variant in this gene had a lower rate of type 1 diabetes.

Obviously, all of this holds tremendous implications, possibly further indicting a maladaptive immune response to a viral trigger as being a potential etiology for diabetes. We don't currently have any antivirals for the viruses implicated, so we can't fend off diabetes that way. Also, this is all for type I diabetes, what is now well-recognized to be a very different disease than type 2 diabetes, for which I have yet to see documented evidence of a known infectious trigger.

All this to say that it is an exciting time to be researching infectious diseases, since they seem to be the common pathway for so many different illnesses. The lesson: Wash your hands.

Eugène Sobngwi, MD, PhD; Siméon Pierre Choukem, MD; Felix Agbalika, MD, MSc; Bertrand Blondeau, PhD; Lila-Sabrina Fetita, MD; Céleste Lebbe, MD, PhD; Doudou Thiam, MD; Pierre Cattan, MD, PhD; Jérôme Larghero, MD, PhD; Fabienne Foufelle, PhD; Pascal F (2008). Ketosis-Prone Type 2 Diabetes Mellitus and Human Herpesvirus 8 Infection in Sub-Saharan Africans JAMA, 299 (23), 2770-2776

Richardson, S., Willcox, A., Bone, A., Foulis, A., & Morgan, N. (2009). The prevalence of enteroviral capsid protein vp1 immunostaining in pancreatic islets in human type 1 diabetes Diabetologia DOI: 10.1007/s00125-009-1276-0

Nejentsev, S., Walker, N., Riches, D., Egholm, M., & Todd, J. (2009). Rare Variants of IFIH1, a Gene Implicated in Antiviral Responses, Protect Against Type 1 Diabetes Science DOI: 10.1126/science.1167728

Tuesday, April 14, 2009

The slow death of inbreeding

As a caveat, I do not profess to be a geneticist, although this study is straightforward enough thatI feel like I understand it well enough to explain it. It's also nice to branch out and read a non-infectious disease based study.

This one, from this month's PLOS Genetics, is interesting, not necessarily for its methodology, but more for some of the conclusions that can be drawn from its results. What the researchers did is take two groups of heterogenous, American people, aged 19-99, performed genome-wide analyses, and quantified the degree of autozygosity within each. By autozygosity, they mean strings of homozygosity, which they took as a surrogate quantifier for degree of consanguinity, or inbreeding, in a population. They also controlled for linkage disequilibrium by getting rid of the SNPs that may have confounded the results. Their results are nicely summed up in the figure below (it's not as blurry if you click on it), which looks at birth date on the x-axis, and on the y, clockwise from upper left, percent of genome in these homozygous runs, number of runs, average length of the runs, and their inbreeding coefficient:

Essentially, the younger their subject, the lower percentage of their genome occupied by these runs of homozygosity, which has, in previous work, been linked to the degree of inbreeding in a population. They make the subsequent statement that the demographic shift over the past century, both from rural to urban and globally, has led to this admixing of gene pools, reducing the amount of inbreeding that's occurring. Since I can't think of any other plausible hypothesis to argue against this conclusion, I have to go along with it. Basically, your grandmother's generation was more inbred than yours, mostly because she didn't travel to the big city after college.

Since consanguinity increases the risk of rare genetic diseases significantly, we may be on the cusp of seeing these diseases disappear. From an evolutionary perspective, this makes perfect sense, as these diseases don't confer any advantage whatsoever. And beyond just reducing the incidence of otherwise rare diseases, this genetic admixing may provide all sorts of differential advantages that we've yet to determine. Another argument for leaving home and traveling. You're doing it for the human race.

Reference: Nalls, M., Simon-Sanchez, J., Gibbs, J., Paisan-Ruiz, C., Bras, J., Tanaka, T., Matarin, M., Scholz, S., Weitz, C., Harris, T., Ferrucci, L., Hardy, J., & Singleton, A. (2009). Measures of Autozygosity in Decline: Globalization, Urbanization, and Its Implications for Medical Genetics PLoS Genetics, 5 (3) DOI: 10.1371/journal.pgen.1000415

Monday, April 13, 2009

A passive approach to influenza

I'm a bit late in the game to this one, since it's been covered in much of the lay press over the past month, but I feel it deserves some further discussion, mostly because of its relevance.

What Jianhua Sui at the Farber here in Boston, with colleagues at the Burnham out in California, and the CDC, (subscription required) have done is potentially develop a new therapy for influenza disease. This has made it around the mainstream press, mostly due to the fears of pandemic illness. Scaling this method up to a pandemic level intervention is another matter entirely, but the science about its utility is outstanding, nonetheless.

Influenza is a notoriously cagey virus, with its two main antigenic determinants, hemagglutinin and neuraminidase, always drifting along, forcing us to reactively change our vaccines every year (that's what the H&N refer to whenever people use words like H5N1, H3N2 and the like). Further, evidence of resistance of our major drugs used to treat the infection is piling up (here and here, for example), making our battle against infuenza a near-Sisyphean struggle, with the especially scary concept of the mutated virus being more fit than the wild-type.

So, what we're left with is trying to completely shift our approach to influenza treatment, focusing on portions of the virus that don't constantly change with the slightest amounts of genetic pressure. That's exactly what Sui's group did: they screened a phage-display library, consisting of about 27 billion unique antibodies, finding a group of antibodies which inhibit, both in vitro and in vivo (mouse model), influenza activity through blocking hemagglutinin activity. These antibodies, on crystallography, bind to a conserved region on the stem region of hemagglutinin, an area which doesn't change so quickly, potentially producing a catch-all group of antibodies effective against every influenza virus. The concept and cost of passively immunizing an entire population to ward off pandemic influenza involves economies of scale that can only be fantasized about at this juncture, but something that deserves further thought after the requisite human studies are performed over the next few years.

One of the lingering questions, that the authors address, is why we haven't figured this out on our own, ie why don't we generate these antibodies after infection, instead of the only transiently useful antibodies that we currently produce? There's no real answer to this, so I'll await further studies before postulating ideas.

It's clear that our current anti-influenza strategies of seasonal vaccinations are only effective in the setting of slowly drifiting virus and significant coverage. In the setting of a significant antigenic shift, further options might be required to avoid pandemic disease, and this option holds definite promise.

Ref: Nature Structural & Molecular Biology 16, 265 - 273 (2009) Published online: 22 February 2009 | doi:10.1038/nsmb.1566

Thursday, April 9, 2009

A sum greater than the equal of its parts

Another PLOS One paper, another malaria paper, another paper from Tanzania, another paper on women's health. I might branch out at some point. Or I might not. It's an exciting time for malaria research in Africa, so there's lots of new stuff coming out.

Malaria in pregnancy is a huge public health problem, with tremendous implications for fetal development as well as maternal health, and every bit of knowledge that we can use in its treatment is vital. Like this paper, looking at combo anti-malarials versus monotherapy, where the researchers randomized pregnant women with mild falciparum malaria to four different treatment regimens and looked therapy failure after four weeks. Knowledge of local resistance patterns is crucial in understanding the implications of the study - these lessons can't necessarily be applied in Asia or Central America, for example, or can't necessarily be applied 5-10 years from now in Tanzania itself. That's why we constantly need studies like this, to prove that we're still doing the right thing when treating this constantly adapting illness, trying to stay one step ahead while prevention efforts continue (see prior blog post).

Beyond the main finding of the combination of anti-malarials being effective, while its two components are individually ineffective, the finding of differing efficacy between study subjects and children reinforces the idea of partial immunity being helpful in fighting off the disease. Children likey need completey different, more intensive, typically artemensin-based, regimens compared to adults in the same community, a striking idea that may be unique to malaria treatment in the world of infectious diseases.

Ref: Mutabingwa TK, Muze K, Ord R, Briceño M, Greenwood BM, et al. (2009) Randomized Trial of Artesunate+Amodiaquine, Sulfadoxine-Pyrimethamine+Amodiaquine, Chlorproguanal-Dapsone and SP for Malaria in Pregnancy in Tanzania. PLoS ONE 4(4): e5138. doi:10.1371/journal.pone.0005138

Thursday, April 2, 2009

HPV - Huge Paradigm Variation

I'm not sure if that title makes sense. Regardless, this is a wonderfully important study, that even though the big blogs will assuredly address it, I need to try and publicize it as much as possible. Published in New England (online only, text version not until the end of the month), these Indian researchers have shifted how we will screen for one of the most deadly diseases on the planet. The evidence against the Pap smear has been mounting for some time, with convention being the only residual argument for it in our changing molecular environment, with effective vaccines and with molecular testing. A developed world study, also published in New England (from my two alma matars), argued that HPV testing has better sensitivity for detecting cervical cancer. Now, in a developing region, we have evidence not just for the arguably-useful outcome of cervical cancer detection, but mortality itself, with HPV testing saving significant numbers of lives compared with traditional methods of cervical cancer screening. Crucially, the patients testing negative for HPV had no deaths in the eight years they were followed, ie the false negative rate was appropriately low.

I'll leave aside the methodology - it appears sound from my assessment. They have powerfully shown that screening women greater than 30, ie at least 15 or 20 years from their first sexual exposure, with HPV testing leads to an appropriate detection rate of cervical cancer. These cases can then be treated, with proven survival benefits, skipping the notoriously subjective Pap smears and visual inspection (especially in cost-limited healthcare settings with large quality variability). The treatment algorithms haven't been fully codified yet - but hopefully this will be worked upon.

They have demonstrated that HPV testing is both reproducible and objective, and something that can be easily scaled up in the developing world. The next steps are to bring down the costs of the tests themselves, something that will hopefully happen as demand goes up and competition arises, as well as governments throwing in their funding dollar in what is clearly an important area of investment. Also, figuring out just how often to screen women is still to be determined - a ten year window may be reasonable, though long-term studies will need to be done to determine that. In addition, making sure that any tests include all of the HPV serotypes, not just the ones covered by the new vaccines, is crucial, as widespread vaccination may lead to serotype switching. What level of screening an HPV-negative women deserves still is under debate, however. Also, screening of young women, ie less than 15 years after their first sexual exposure, with their extremely low incidence of cervical cancer may be antiquated.

The accompanying editorial makes the good point that this may be slow to disseminate in places like the USA, where Pap smears form the basis of primary care for women in this age group, and that cytology could be used as a secondary screen for HPV positive women. Regardless, this is a huge leap forward for cervical cancer screening for the entire planet, and may be one more step towards making the routine Pap smear a thing of the past.

Tuesday, March 31, 2009


The classic model for malaria is rural and tropical. That model is changing with recent demographic and ecological shifts, as malaria is becoming more and more an urban disease, with the expected alteration in required prevention tactics. Unfortunately, the urban lifestyle dictates less time spent at home and more time out-and-about in the city, meaning traditional and proven control mechanisms such as insecticide treated nets lose some of their effectiveness. In addition, traditional agricultural development that has been shown to be effective in curtailing malaria propagation is less relevant in urban centres (I suggest Randall Packard's great book on the history of malaria for more information on this). Therefore, new techniques are mandatory, mostly through mosquito control. While we wait for genetically engineered mosquitoes, good old fashioned mosquito killers will have to do.

The larvicides used in this study (Geissbühler Y, Kannady K, Chaki PP, Emidi B, Govella NJ, et al. (2009) Microbial Larvicide Application by a Large-Scale, Community-Based Program Reduces Malaria Infection Prevalence in Urban Dar Es Salaam, Tanzania. PLoS ONE 4(3): e5107. doi:10.1371/journal.pone.0005107) published this week in PLOS One, are bacteria distributed in pellets that larval mosquitoes ingest. These bacteria release toxins that are toxic to the mosquito digestive tract, resulting in mosquito death. The pellets are effective for a few weeks after distribution, and there has been no documented side-effect of human danger from their dissemination. One of the challenges is the development of larval resistance to their toxins, something not seen yet, but potentially present in the future.

The researchers showed a fairly significant decrease in malaria incidence, and a decreased number of mosquito bites in general (with the rigorous bites/night methodology used by frequently-tested volunteers). A nice population level intervention, though it was relatively top-heavy with staff and infrastructure - for an urban centre, however, the cost-effectiveness is probably reasonable, given the huge economic effects of malaria incidence upon a population. The next step is to replicate this in another city - perhaps in Asia where the mosquitoes are slightly different, to see if this can be effectively scaled up for global dissemination.

Friday, March 27, 2009

Babies with bad guts

The first post of many, hopefully.

I'll try and focus upon research that may not otherwise get the press that it deserves, either because of the obscurity of the topic or the location of the researchers. Here's an interesting article performed by researchers in Kolkata, randomizing premature babies to prophylactic probiotics or placebo, and finding a fairly significant response rate with the outcomes of necrotizing enterocolitis, hospital stay, and time to full gut feeding. Not revolutionary stuff, since prior studies have backed this up, including a well-performed Cochrane analysis, which showed that by giving little babies probiotics, you reduce the incidence of this serious gut disease. This has been slow to disseminate into practice, however, since people seem to be still afraid of giving little babies, with their immature little immune systems, more microorganisms. None of the data says that this is a concern, with no significantly increased rates of new blood infections or other serious adverse events in the babies that received the probiotic.

What's more notable than just the direct clinical application of this study, however, is just how crucial our endogenous bacterial flora is in our physiology. Premature infants are premature in everything, including the establishment of an appropriate microbial environment, and by artificially inducing something akin to that, as the researchers have done through providing organisms that are commensal in breast milk, they have reduced the incidence of a disease. Bacteria have long gotten a bad rap, and only recently their multi-varied function in the developing infant has been realized: from digestion to vitamin production to immune regulation
to many more, emphasizing the complex host-microbial relationship. Some exciting stuff is being performed looking at the long-term incidence of disease and the neonatal gut flora, and most exciting is the human microbiome project, with the potential to seed the gut with specific disease-preventing/treating bacteria and add even more complexity to the gene-environment interactability discussions. Exciting times.

An initiation

I've tried blogging numerous times before. It's fairly straightforward, though my attention span doesn't seem to be conducive to maintaining and sustaining a coherent theme to one specific blog. Hence, my blogger profile is littered with the corpses of blogs-that-were-but-are-no-longer-active, although this time will be different (I swear). Instead of nonsensical ramblings about the state of the world, this will be a structured, weekly compilation of published scientific literature that I find the most interesting and relevant to the world at large. Yes, there will be digressions into discussing the latest Ig-Nobel worthy feline gender identity research, but hopefully I'll maintain my focus on the studies that really matter, leaning towards my tripartite passion of infectious diseases, child health, and critical care.

Today, as an intro, one of my favourite TED talks of the past few years, not necessarily because of its debatable content, but moreso because of its effective use of data visualization to convey a complicated message.