» Smallpox

The following is a transcript of a comment by Dr. Tom Mack, University of Southern California, at Centers for Disease Control and Prevention's (CDC) "Public Forum on Smallpox":

As you probably know, I'm at the University of Southern California School of Medicine. I've been out of the smallpox game for roughly 40 years. My credentials include probably spending more time working up population-based outbreaks of smallpox than virtually anybody ever has. We spent three years in Sheikhupura district in Pakistan and worked up 121 outbreaks, which we estimated were roughly 85 percent to 90 percent of all the smallpox that occurred in that population of a million or so people.

And the experience contrasts somewhat with a lot of the other series because population-based outbreaks include small outbreaks that never result in any hospitalization, and individual importations, which never result in any cases coming to the attention of authorities.

So in the Pakistan study, this is roughly a situation where more than a quarter of the people were unvaccinated. It tends to be villages of from two to 20 to 100 crowded compounds, 1,000 to 5,000 people. Any given village received an importation maybe once every 10 or 15 years, so these were people who were not familiar with everyday smallpox. And in essence, there was really no medical or public health care, and there are a variety of political and historical reasons for that, which we can go into, but the import of it is that there really was no intervention.

You heard several references to my review of the European experience. I'd like to reiterate that this was -- these experiences were in essentially susceptible populations with physicians who were unfamiliar with the disease, media and communication skills much less than today, and the standard of living actually substantially less than today in both Europe and America. So in my opinion, the propensity for spread in both these circumstances is substantially greater than it would be in the United States today.

You're going to hear more about vaccinia. I'm not going to spend any time on that. I just want to point out the last word in this slide, which is VIG. I haven't heard that phrase mentioned today. To me, it should be an extremely important consideration in all of your deliberations because in the absence of VIG, any extensive vaccination would be extremely dangerous.

I'll try and skip data slides because you've seen many of them already. This just reiterates the effectiveness of past vaccination, and in this case it demonstrates that the severity of disease was affected.

This is the study that was previously referred to by the Russian gentleman who tried to vaccinate people who'd previously had smallpox, demonstrating that the history of severity was an important determinant of whether or not he could get vaccination takes, irrespective of the interval since the case occurred.

O.k., the trade-off is with smallpox, and I'd just like to point out that not only is it a nasty syndrome, but the case fatality is probably less than is usually advertised. And the reason for that is that most series are heavily loaded with children. If you look at the age-specific case-fatality rate, it's much lower among adults. And so I would estimate that if we had an importation today in the adult population, the case fatality would probably be around 10 percent to 15 percent.

It does have a truly terrifying pathognomonic appearance, and that's one of the characteristics that would make control much easier. Again, as has been mentioned, there's acute illness during a brief period of infectiousness. There are no reservoirs or vectors. There is a finite half-life in the environment.

And most importantly, there's a big -- one to two -- one to three-week interval between generations in which activity for surveillance and containment takes place. And by and large, transmission within social limits is what occurs, not within the population at large. And these, by and large, cannot be sustained. In fact, were there no smallpox eradication program, my guess is that smallpox would have died out anyway, it just would have taken a lot longer.

Now a few slides to show you what it actually looks like. That's hemorrhagic smallpox. This lady was actually not vaccinated, she just has sparse disease. But you can see that the characteristics of the lesions are just the same.

This is a girl at three days of rash. I don't think anybody could pick up that that is smallpox without an awful lot of experience. This is the same girl at seven days.

This is also a man at three days. Unfortunately, I didn't have a slide of him, but he died with very rapid confluent smallpox, and you can detect that it's going to be confluent from his appearance here.

I'll go over these slides because they've been shown before. In 27 percent of the cases in Sheikhupura, there was no transmission at all. Another 37 percent, only one generation. Now we're talking about a place where there really was no care given. The mean length of the outbreaks was six weeks. That's roughly three-plus generations, so a few of the outbreaks were longer. We could detect the source.

Virtually all the people we could identify as introducers, even though sometimes we came upon the outbreaks substantially late. In other words, most people knew where they got smallpox. It wasn't a matter of their having gotten it on a train or gotten it in an unknown place.

The top figure here shows the distribution of cases in the same compound as an introducer, showing essentially the incubation period variation, and it does correspond to what's been known before, one to three weeks. The lower one shows the distribution of cases in other compounds.

I would point out that when one is looking at attack rates, they're always confounded by the nature of the social arrangement. A compound in West Pakistan is very different from a compound in West Africa. The people are much more closely in contact, and so the attack rates here were much higher than they were in West Africa, and I daresay also in Madras because the living arrangements are very different. And the definition of what constitutes a unit for study is very different.

This has been referred to in the past. Twenty-seven percent, again, no transmission. Thirty-five percent, only one or two indigenous generations. Even with the hospitals, no more than six generations. The largest generation was 20 cases.

This is an illustration of the effect of temperature and humidity on the occurrence of disease. This represents the seasonal distribution in Sheikhupura, almost the same figure we derived from seasonal distribution in east Bengal. During the dry season, the cases are much more effectively transmitted than during the wet season.

And in fact, it's not just a function of population movement, but it's actually a function of virus survival. If you just look at the last two figure on this graph, there are three times as much effectiveness of transmission to other compounds in the period when the increase in incidence was occurring than when it was decreasing.

O.k. What do we expect if there were a terrorist introduction? I would expect a small number of cases. I don't think suicide dissemination is a very likely possibility because of the severity of the disease. I think that airborne spread would be relatively inefficient and I don't think very many cases would occur, and that's just giving you my personal opinion.

The danger would be from release within a close space, like an airplane. Then there might be several -- a substantial number of cases, but they would all share a common experience and probably could be identified. Cases would be florid because we're an immunosusceptible population, by and large. People would be aware of exposure after the initial diagnosis, and I think dissemination from the individual cases would probably be relatively limited.

The key to any introduction would be, as has been mentioned, surveillance. I think initial recognition would be the most important single factor. Identification and follow-up of contacts, obviously. Isolation of known and probable cases, preventing admission to hospitals and opening separate facilities, and then vaccination of likely contacts.

Initial recognition, to me, awareness of the possibility of the disease, is vastly more important than the details of how to distinguish it from chickenpox. I don't like to see posters with lots of fine print. I like to see a poster with a really big picture, and that would make people aware that the possibility exists. And when they saw that, they'd run to the books and they'd learn what they could otherwise see on posters. Large subtleties will seep through afterwards, just as they did with the anthrax situation.

And by large photographs, this is the kind of photograph that I would like to see on a big poster. These are the classic lesions. That could not be any other disease.

Now very early, it's difficult. But after a few days, there's not going to be much likelihood of error. There will be the occasional case modified by vaccination. But if there's only one case, what worries? There'll be subsequent cases and they'll be much more likely to be diagnosed. So we may miss some if it were to occur, but we won't miss very many.

That's the flat variety that was referred to earlier. Contact identification. Personnel. We don't need vaccination, we need personnel. If smallpox were to come to Los Angeles tomorrow, the more cases we'd expect, the more people we need prepared, and that -- those people may come from San Francisco, they might come from Atlanta, they might come from Michigan, under ideal circumstances.

I would like to think of the firefighting as a model for how to deal with a smallpox outbreak. People might be prepared in every locality, and then be gathered together when necessary. The more public exposure, the more people are needed, for obvious reasons. Availability of protected personnel to me is vastly important, and that would mean field epidemiologists, lab people and care providers, designated people. And I would suggest that older and foreign MD's who were previously vaccinated ought to be given priority. Multi-locality Federal cooperation is really advantageous.

Most important determinant to the eventual number of cases is whether or not somebody gets put in the hospital. And everything should be done to prevent that, and the most important thing is initial recognition. That depends on the state of alertness and familiarity with the possibility of the syndrome. And a dedicated facility need not be large, but better small and agreed-upon than large and contentious. I don't know whether you people have had such discussions in your localities, but we certainly have.

Populations requiring separate vaccination policy. See, I didn't know those three questions, but I anticipated them. Vaccinating those expected to implement control, those known exposed to a case or an exposed person, those not so exposed but at risk of work place exposure, and members of the community at large.

With respect to the first, I think this is very essential that there be designated individuals who are vaccinated in advance, with VIG available and with screening for those at risk for complications. Oops, I think I missed one. That's all right. Yes, that was my point.

This is the people who are actually exposed. Now I was asked to speak about post-exposure vaccination a few minutes ago. There's not much to say about it. I can give you the little bit of data that I have. It isn't very good data. I expect that post-exposure vaccination does make a difference. I don't know exactly, on a day-specific basis, how that difference changes. I would certainly want to be vaccinated myself, and I would want to vaccinate my relatives. I would also think about passive immunization and chemotherapy, of which I know nothing -- the latter, at least. But we would do whatever we could.

We're going to have to expect, if there is an importation, that there are going to be people we do not identify who have already been exposed, so we'd better prepare for it.

These are figures that have already been shown -- no, that's not. That's not true, sorry. The last two columns compare post-exposure to no vaccination. You can talk yourself into there being a difference, if you wish. My guess is there is, but I couldn't convince any biostatistician of it. Similarly, nine out of 19 who got post-exposure vaccination were affected, compared to one out of three. Can't make a big case out of that. And by the same token, 12 out of 16 versus 26 out of 27. If you put all these numbers together, you might or might not get statistical significance. They would be heavily confounded by a variety of circumstances, which are not under control, so I wouldn't want to say we have strong evidence that it works, but it should be done anyway.

I have the opinion that doctors and emergency room workers should not be vaccinated a priori, as a category. I think that is true because the likelihood of their being exposed, even under circumstances of importation, is very, very small. And I also think that that will eventually become mass vaccination, whether we want it to be or not. They will be concerned about their families. There will be people making decisions who have not thought through the risk issues.

Policemen and firemen and everybody else who potentially might be exposed under a contingency will demand equal treatment and I don't think it'll work.

Unexposed community members have negligible risk. There is a substantial risk from a vaccine, as you'll hear in a moment. It is the single most dangerous live vaccine. We would still need to vaccinate and identify contacts. We would need personnel and resources for surveillance rather than mass vaccination. That protection will not be maintained. It will gradually wane and we'll have to do it again and again.

The informed consent that you would have to prepare to vaccinate somebody in the public, if it's honest, would have to say that the dangers would exceed the benefits. And even if you fudged those words in such a way that you were happy and thought it would be convincing, an awful lot of people who ultimately might be exposed would not be convinced. You'd have to go back again anyway. So I don't think it would work and I don't think it would be beneficial.

If people are worried about endemic smallpox, it disappeared from this country not because of our mass herd immunity. It disappeared because of our economic development. And that's why it disappeared from Europe and many other countries, and it will not be sustained here, even if there were several importations, I'm sure. It's not from universal vaccination.

So if I were the New York health czar, knowing a case would get on the subway, I would rather have the money to prepare field workers than to give mass vaccination. The first unnecessary death from a vaccination complication would result in more, not less, smallpox transmission because people who needed the vaccination under that circumstance would refuse it. The presence of partial herd immunity would not lessen work and might lead to complacency.

So my views on the three questions are obvious. I would choose option one for the first one. I would choose option two for the second one. And I would emphasize the inclusion of local people because CDC cannot respond quickly enough, and there will become -- when the difference between the second or third post-exposure day and the sixth through the seventh post-exposure day might be important. And under option -- number three, surveillance, surveillance, surveillance. It's not ring vaccination, it's surveillance. Vaccination is a subsidiary issue. Thank you very much.

© Copyright 1997-2002 Dr. Joseph Mercola. All Rights Reserved.