If you are in the "older age group" (as those of us in that group prefer to be called) you are at increased risk of dying from seasonal influenza (pandemic strains seem to target the young), but you are also less likely to be helped by a flu shot because you don't mount as fast and effective an immune response. At least that's what we think on the basis of current evidence. Like everything else about flu, it's subject to change.
Like the idea that you have a 48 hour window for the use of antiviral neuraminidase inhibitors like tamiflu. After 48 hours, we believed, they don't do any good. Maybe that's true, but some new evidence suggests it isn't:
People stricken with a case of flu so severe it sent them to hospital were five times less likely to die if they were put on antiviral therapy compared to those who didn't receive one of these drugs, the study found.The findings could alter the way doctors treat hospitalized flu patients, experts who were not involved with the study said.
"I think and talk about flu a lot. And this is going to change what I say," said Dr. Anne Moscona, an infectious diseases expert at Weill Cornell Medical Center in New York City.
[snip]
"It's clear from this experience now that there seems to be benefit even with later treatment," Dr. Frederick Hayden, an antiviral expert with the World Health Organization's global influenza program, said from Geneva.
Hayden said the study will "add to the body of evidence . . . that hospitalized patients warrant therapy if there's evidence for ongoing (virus) replication."
The study evaluated what happened to 327 adults who were hospitalized with influenza in a network of south-central Ontario hospitals during the 2004-05 and 2005-06 flu seasons. The work was led by researchers at Mount Sinai Hospital and the University Health Network, both in Toronto. (Helen Branswell, Canadian Press)
Maybe it's clear to dr. Hayden, but I don't know that yet. Unfortunately this otherwise informative article doesn't say where the work was published or otherwise described, so I can't say much more about its methods. The study subjects as a whole were high risk because of their age and other medical conditions. All were said to have tested positive for infection with influenza A, but it was not a randomized trial because of the ethical issue of withholding a drug thought effective, so it remains possible those treated with antivirals (32%) at the discretion of their physicians were different in important ways that affected their survival from those not treated. Without the details it is hard to say more, and it is likely even with the details we will have to consider this another data point, although an important one.
If you are in the "older age group" it is not bad news and these days I am glad to settle for news that's not bad, even if I don't know yet if it is really good.
- Log in to post comments
I would be most interested in this article as I recently did a search looking for mortality data regarding the use of oseltamivir. I couldn't find any such data.
It's in CID.
from the branswell piece:
It will be published in the Dec. 15 issue of Clinical Infectious Diseases.
Lead author Dr. Allison McGeer
there is also a major confounder: many but not all had been vaccinated. it is possible that vaccination does not prevent disease, but does prevent death.
This is an interesting article
It suggests that use of oseltamivir might be helpful in those with seasonal influenza at high risk for death irrespective of whether treatment is initiated within the 48-hour window.
It tells use nothing about the efficacy of oseltamivir in those with H5N1 pandemic influenza. Many who died and some who survived H5N1 were given oseltamivir during the course of their illness. Others who succumb to it and some who survived never received this drug.
What is necessary is a double blind randomized clinical trial of those ill with H5N1 that compares usual care plus or minus oseltamivir treatment without regard to timing of initiation of therapy. The data from a properly powered trial of this type would inform us on the efficacy of this treatment during the course of pandemic flu.
Grattan Woodson, MD
Gratt: I think there are many problems with your suggestion. One is whether it is ethical. In addition, an RCT is not likely to answer the question. RCTs are much harder to do than most appreciate and having one that is "properly powered" even more so. Consider mammography. People are still arguing about it even though there have been 8 RCTs. And why double blind? Not necessary.
I agree that an RTC would be difficult but not impossible to accomplish.
With regard to ethics in clinical trials, it is very important to consider the risk to the individual and the benefit to society of a study before an IRB can decide whether or not to approve a RCT. The fact is we have no idea if use of oseltamivir at any stage during the course of H5N1 illness has any benefit whatsoever. We do know there are serious risks of using this drug in children for seasonal influenza. The original RCTs of oseltamivir showed no significant benefit to the patient if the agent was administered later than 48-hours after the onset of seasonal flu symptoms. It is well appreciated that the logistics of getting oseltamivir into the hands of those with H5N1 during a pandemic within the 48-hour treatment window is not something that is going to occur with regularity.
Given this fact and what is a stake here, human investigation review boards would, in MHO be more likely to approve a study like this than not. The key ethical consideration is that "usual care" be the same for all participants. The fact that "usual care" is likely to be very different in different places is a potential confounder as the power of usual care may turn out to be a far better predictor of outcomes that whether or not the subject received oseltamivir. So the extent of the differences in usual care could introduce and new independent variable in addition to the study drug. This factor can be sorted out from the study drug effect through statistical analysis and should not in and of itself be a reason for the RTC results to be invalid.
Other issues include the likely fact that the patient demographics will differ significantly with regard to age and comorbid conditions. These confounders simply cannot be overcome today especially with the growing but still small number of new cases.
Of course we cannot predict how or if H5N1 will become pandemic and if it did what path it would follow but just suppose what we see a rising rate of new cases consistent with WHO Phase 5. This phase might percolate along for much longer than some expect. The number of cases in Europe, Asia, and Africa in this scenario may reach into the thousands rather quickly, a number that would certainly be adequate to power the proposed study even with the recognized confounders.
It would really depend on how significant the treatment effect of the study drug was when used in these subjects. If large, then this would be an important finding. Even if the null hypothesis is confirmed in an adequately powered trial, there might still be meaningful data that could be harvested from the study regarding early verses late initiation of oseltamivir treatment.
Into the mix comes the article in question that disputes the RCTs data leading the registration oseltamivir for seasonal flu. Since the quality of the recent report is lower than the RTC, its findings should not be accepted until repeated using a better design. Even more important is that we resist the urge to extrapolate these recent data to the treatment of H5N1.
We need to know the answers to these questions and also to the dose and length or treatment required for treatment of H5N1. Clinical medicine has spent the last 20 years fighting our way out of the smoky forest of empirical treatment into the clear field of evidenced based medicine. Here we stand on the precipice of a great human catastrophe where use of EBM could really make a big contribution to conventional care of hundreds of millions of people. Will we use our new scientific tools or shake our hands with why we can't do this or that? We at least need to try in my view. At some point we would get the answer maybe not as soon as we would like but almost certainly at a point in the pandemic when this knowledge would be of critical value in the care of a whole lot of very sick folks.
Grattan Woodson, MD
Gratt: I support evidence based medicine but I am also clear eyed about its abilities. RCTs rarely answer the questions they set out to answer because if they answer were clear enough for a single or even several RCTs to answer it we wouldn't need the RCT. Well designed observational studies ae often superior to fair to middling RCTs and in this case the possibility of an RCT (since it would have to be done during a pandemic to be adequately powered) is virtually nil, IMHO, and if we had one it wouldn't be determinative. Take a look at the mammography RCTs (and the subsequent controversy over them in the Lancet meta-analyses by the Danish group in 2002) if you want to see what can, and will, happen. Setting up an RCT as the gold standard here is neither realistic nor correct in my view. It is neither feasible nor required, as you imply. Yes, it would be good to have some RCT data points if there were any or we could get them, but they would just be data points and we aren't going to get them. The benefit to society is not sufficient for most IRBs to allow patients to be put at risk, but that is not the main point.