An HIV/AIDS "skeptic" questions my honesty and decency...

Dean Esmay, who likes to present himself to his readers as a hard-nosed "skeptic," but is in actuality a rather credulous fellow, at least when it comes to HIV "dissidents" and seemingly not understanding why it's such a bad idea to teach "intelligent design" creationism in the science classroom in public schools, is not happy with me, not happy with me at all. On Saturday, as I was about to sit down to watch the Michigan-Ohio State game, I noticed an influx of referrals here from his blog. Given that I hadn't recalled seeing referrals here from Dean before, I checked it out his blog and, amidst a self-congratulatory tirade against bloggers who considered the Eliza Jane Scovill case a tragic example of what can happen to a child when her parents believe won't accept the science showing that HIV causes AIDS, found this comment regarding the case of Eliza Jane, the daughter of HIV denialist Christine Maggiore, who died in May and whose autopsy showed that she had AIDS-associated pneumonia:

I wonder if the Suicide Girls or Orac Knows will show any class, decency or honesty on the subject now that we know the truth? Or will they keep to their lynch-mob prejudices?

Geez. I don't know what brought this on. I've never questioned Dean's honesty. I do think he's prone to excessive credulity when it comes to HIV "skepticism" and that HIV denialism has the potential to do great harm in terms of controlling the spread of AIDS, but I've never impugned Dean's honesty or decency.

OK, Dean. You can trumpet your pseudoscientific nonsense about HIV and AIDS to your readers all you like. I don't really care. You can say that I'm mistaken. You can imply that I lack class. You can even call me a flaming idiot and moron. I wouldn't really care all that much. You can say I'm butt-ugly and smell bad. Water off a duck's back. But question my honesty or decency, and, dude, the gloves come off, making today a very good day indeed for Orac to direct some of his characteristic Respectful Insolence™ your way.

So what brought on Dean's little tirade against others bloggers (and me apparently as an afterthought)? Christine Maggiore, the HIV-positive mother and high profile HIV/AIDS denialist who refused to take AZT to prevent maternal-fetal transmission of the virus, or even to test her child for HIV, has gotten fellow HIV denialist Dr. Mohammed Al-Bayati to look at the L.A. County Coroner's report for her daughter's autopsy. Not surprisingly, Dr. Al-Bayati is claiming that the coroner got it all wrong and that Eliza Jane didn't die of AIDS-related pneumonia as the report concluded, leading Dean to go into spasms of self-righteous rage and claim "that the L.A. County coroner and the Los Angeles Times were and are guilty of a political diagnosis in order to grandstand."

Well, knock me over with a feather! Who could possibly have seen that one coming? The only thing that surprised me is that it took so long to dig up someone to "refute" the coroner's report.

One thing that struck me right away is that Dean failed to mention other medical bloggers who agreed with my position and said so in their blog, bloggers such as Gordon's Notes, Red State Moron, SupportVaccination.org, or even Dr. Trent McBride, a pathology resident who blogs on Catallarchy and who also pointed out that finding Pneumocystis carinii by silver stain on autopsy histology of the lungs of someone not having AIDS or other significant immunosuppression is exceedingly rare. (I wonder if Dean considers these other bloggers writing on this case to be lacking in honesty as well.) Another thing that struck me is that, in the comments Dean whined about ad hominem attacks against his expert (Dr. Al-Bayati), after having himself launched an ad hominem attack on Richard Bennett (whom, I conceded, had once taken what I considered to be a cheap shot at Dean over an admission Dean had made) and others. No doubt Dean fails to see the irony. Apparently ad hominems are OK only as long as it is Dean--who has on occasion been known to refer to his critics by epithets like "sad little losers"--launching the ad hominem.

Nonetheless, at the risk of my being accused of retreating to the "last refuge of the pseudoscientist" (as Dean himself likes to put it), the ad hominem attack, it is nonetheless necessary to take a brief look at Dr. Al-Bayati before addressing the substance of report itself. (Don't worry, I'll get to the substance soon enough.) Given that Dean seems to like to question the motivations of those who don't buy into his "HIV dissident" line and to argue from authority a lot (he is quite enamored of his friend Peter Duesberg's credentials and those of another prominent AIDS denialist, Harvey Bialy and is not at all shy about waving them in front of him--metaphorically speaking--like a talisman to ward off attacks against his pseudoscientific posturing), I consider it entirely appropriate to examine this particular "expert" and his qualifications and motivations before going on to discuss the contents of his report.

So who is Mohammed Ali Al-Bayati, PhD, DABT, DABVT? He represents himself as a toxicologist and "pathologist." However, most pathologists who deal with HIV are MD's. So what kind of pathologist is he? "DABVT" stands for Diplomate, American Board of Veterinary Toxicology; so basically he's a veterinary pathologist and toxicologist. Whether that means he's qualified to evaluate postmortem findings in AIDS, I don't know. He has a few papers published in the peer-reviewed medical literature, but none of them concern HIV; so his publication history doesn't help me evaluate him. He does, however, have a very obvious and undeniable bias. Dean will no doubt characterize it as an ad hominem attack to point out Dr. Al-Bayati's bias, but it's not at all inappropriate in this case to mention his prominent listing on the infamous Virus Myth website or his book Get All the Facts: HIV Does Not Cause AIDS. Also, Dr. Al-Bayati runs a company called Toxi-Health International, which, according to its website, provides expert witness services and "can evaluate the health effect resulting from acute and chronic exposure" to various agents," including medication reactions, adverse reactions to vaccines (shades of the Geiers!), pesticides, and a variety of other compounds. No doubt Dean will lambaste me for even mentioning such things. Tough. If Dean considers it not to be an ad hominem attack to blithely accuse the L.A. County Coroner and L.A. Times without evidence of making "a political diagnosis in order to grandstand," I consider it acceptable to point out an obvious bias in the source Dean chooses to use to make the case that the autopsy findings were incorrect.

As I've said before, as long as you're not just lobbing insults at your opponent, ad hominem arguments are not always inappropriate, particularly when they point out a clear bias that any person evaluating both sides of an argument should be aware of. However, they are not sufficient. The substance of an argument must be addressed as well. So, without further ado, though, let's get to the meat of the matter, shall we? Unfortunately, Dean provided no link to the original coroner's report (apparently it is not publicly available); so I shall have to rely on Dr. Al-Bayati's characterization of the report. Fortunately, that's enough to sink him.

The first thing I noticed reading Dr. Al-Bayati's report is that it appears not to be intended for a scientific audience. (Perhaps this is why Dean found it so compelling, assuming he actually read the whole thing.) Instead, it is constructed more like a legal document designed to cast "reasonable" doubt on the coroner's conclusion that AIDS-related pneumonia was the cause of Eliza Jane's death, rather than actually putting together a coherent case for an alternate explanation. Without ever having seen the body or any of the raw data from the autopsy, Dr. Al-Bayati dives right in. One particularly egregious example of his style is that he lambastes the coroner for not testing for a certain virus (more about this below), and then confidently concludes that Eliza Jane had that virus and not HIV, even though he couldn't possibly make such a conclusion without the results of the very test that he criticizes the coroner for not having done in the first place! Before I myself dive in, however, let's summarize the autopsy conclusions as Dr. Al-Bayati reports them:

1. Pneumocystis carinii was found in Eliza Jane's lungs by Gomori methenamine silver staining in association with pink foamy casts in the alveoli. The lungs were also edematous (water-logged).
   
2. Eliza Jane was mildly neutropenic (low neutrophil--a type of white blood cell--count) and profoundly anemic (low red blood cell count)
   
3. Eliza Jane's brain contained throughout its white matter with relative sparing of cortex a number of variable-sized microglial nodules characterized by multinucleate giant cells associated with moderate pallor and myelination, occasional macrophages, and and angiocentric pattern. These lesions stained positive by immunohistochemistry (IHC) for the HIV core p24 protein, a finding consistent with HIV encephalitis.
   
4. There was atrophy of the spleen and thymus
5. There was enlargement of the liver with fatty infiltrate of the cells (steatosis) and ascites
   

Dr. Al-Bayati then tries to "refute" each of these findings, using a variety of handwaving techniques and "might have beens" that truly astound. One thing that puzzled me, though, was why he thought it so important to refute the finding that HIV was present at all. After all, if, as Dr. Al-Bayati clearly believes very strongly, HIV does not cause AIDS, then why not just come right out and argue that in his report? Why not just argue that AIDS couldn't possibly have killed Eliza Jane and that the HIV protein detected in her brain was a red herring? Heck, why doesn't Dean just argue that?

But I digress.

Dr. Al-Bayati concedes that P. carinii, an AIDS-defining organism, was present in Eliza Jane's lungs but tries to wave this finding away by pointing out that there was not a "pneumonia" because no inflammation was observed, citing a definition in a pathology textbook (a technique not unlike arguing about technical words using dictionary definitions). He repeats this again and again ad nauseam. He also states that P. carinii is ubiquitous, only causing disease in immunosuppressed patients. There are couple of problems with these arguments. First, immunosuppressed AIDS patients tend not to be able to mount a very effective inflammatory response to infection. Indeed, it has been noted that, in HIV infection, PCP pneumonia provokes fewer inflammatory cells and that PCP is worse in patients immunosuppressed by other causes as their immune system recovers and starts attacking the organism, causing inflammation. (That's one reason why the chest X-ray findings and physical exam findings can be so variable.) The one argument Dr. Al-Bayati makes in this context that isn't totally off the wall is that PCP can occur due to immunosuppression from other causes, and he cites several references that show that PCP can occur in people without HIV if they are immunosuppressed for other reasons. Of course, this line of argument totally begs the question of what the cause of this Eliza Jane's profound immunosuppression was in the first place if it wasn't HIV infection. Second, as Dr. McBride pointed out, for P. carinii to be detected in routine tissue samples at autopsy, there have to be a lot of organisms there. In immunocompetent individuals, there simply aren't enough bugs to show up on silver stain. Given that the HIV protein detected in the brain implicates an obvious cause for the immunosuppression that led to the presence of so much P. carinii in Eliza Jane's lungs, it's hard not to conclude that Eliza Jane had AIDS-associated PCP. Dr. Al-Bayati clearly realized that he had to try to throw doubt on that finding.

And how does he try to do that? Disputing the findings of an experienced neuropathologist, Dr. Maurice A. Verity of UCLA, who examined the sections of Eliza Jane's brain, Dr. Al-Bayati argues that the brain lesions seen are nonspecific and that the finding of the HIV p24 protein on IHC must have been a false-positive. He points out a paper from 1992 indicating a high level of false positivity of this test in the presence of inflammation. Tellingly, however, despite listing the numbers and types of tissues stained in the study (which included only 3 brains from HIV-positive patients and one brain without HIV), he does not cite the percentage of false positive results reported in the paper, only that it is "common." (I couldn't get the paper online to check myself, because the online archives only go back to 2000.) The problem with this line of argument is that it's not enough just to say that this "might" have been a false positive using references that, being 13 years old, may not even be relevant to how IHC for HIV proteins is done today. He has to show compelling reason that it was, rather than hand-waving and saying that some combination of a viral infection and/or an allergic reaction to amoxicillin caused this (see below). Even Dr. Al-Bayati appears to realize this shortcoming.

And that's where his strangest argument of all comes in.

There's a saying in medicine that, when you hear hoofbeats you don't look for zebras. (A zebra is medical slang for a rare or highly unlikely diagnosis.) Yes, occasionally it you will find a zebra, but the vast majority of the time you will not. Consequently, when one hears hoofbeats from a tragic case of a dead child of an HIV-positive mother who was found to have profound anemia, PCP, and encephalitic lesions with HIV proteins detected in them, by far the most likely diagnosis is AIDS. Indeed, in the differential diagnosis, the first ten diagnoses in the differential would be AIDS, AIDS, AIDS, AIDS, AIDS, AIDS, AIDS, AIDS, AIDS, and then--very far down the line in probabilities--everything else. Given this, it's not surprising that, in his rebuttal, Dr. Al-Bayati hears not one, but at least two zebras approaching.

Which zebras, though? The first one is erythrocytic aplastic crisis due to infection with parvovirus B19 (PVB19, the virus mentioned above), of course! Shouldn't it have been obvious? He bases this speculation on Eliza Jane's anemia, encephalitis, upper respiratory infection, and atrophy of the spleen and thymus. I'll give him props, though. This is a clever gambit, because this particular virus is common enough that a significant percentage of children have been exposed to it, meaning that there's a reasonable probability that antibodies to it would be found if looked for, whether it was this virus that actually caused Eliza Jane's death or not. Let's look at this claim a little more closely.

PVB19 is a parvovirus that is fairly common and can cause upper respiratory infections, erythema infectiosum, arthritis and arthralgias, and transient aplastic crisis. Dr. Al-Bayati makes much of the ability of this virus to cause anemia by transiently suppressing the progenitor cells that develop into red blood cells and blames infection with this virus for Eliza Jane's profound anemia. He also attributes Eliza Jane's encephalitis and bone marrow atrophy to infection with this virus. However, he neglects the observation that PVB19 is rarely much of a problem in healthy individuals. Severe anemia secondary to PVB19 usually only occurs in patients with a pre-existing anemia or pre-existing destruction of red blood cells and who therefore require a high level of reticulocyte production to keep their blood counts up. Conditions in which this can be a problem include chronic hemolytic anemias, sickle cell anemia, thalassemia, acute hemorrhage, and iron deficiency anemia, all of which cause red blood cell loss requiring replacement. Also, severe infections of the bone marrow with PVB19 causing aplastic anemia are rare aside from patients with pre-existing immunosuppression, such as transplant recipients, patients with malignancy, and, of course, patients with HIV. In such patients, PVB19 infection can result in severe, prolonged, recurrent, or even permanent anemia. Dr. Al-Bayati does point out a couple of case reports of a aplastic anemia due to PVB19 in immunocompetent individuals. As far as I can find, there are no decent-sized series reported, however, only a few isolated case reports. This paucity of reports that PVB19 is a fairly rare cause of aplastic anemia in healthy individuals. It is, of course, possible that PVB19 infection caused Eliza Jane's serious anemia (that's where "reasonable doubt" comes in). However, even if it had, given what is known about this strain of parvovirus, a far more likely explanation would be that the virus caused Eliza Jane's anemia because of immunosuppression secondary to her HIV infection. Indeed, case reports even suggest that anemia in AIDS patients due to PVB19 will improve with antiretroviral therapy.

The second zebra is a severe amoxicillin hypersensitivity. (An amoxicillin allergic reaction in and of itself is not a zebra, but it is in the context of this particular case.) Never mind that Eliza Jane had, as was emphasized in the report, never been exposed to antibiotics before and that her clinical course does not seem consistent with a hypersensitivity reaction. Never mind that there was no eosinophilia, no urticaria, or no other stigmata of an allergic reaction mentioned in Dr. Al-Bayati's report (and you can bet that, had any of them been there, he would have mentioned them prominently). Amoxicillin hypersensitivity was also blamed for Eliza Jane's steatosis. It is true that amoxicillin-clavulanate has been associated with hepatocellular, cholestatic, granulomatous, or focal destructive cholangiopathy (all types of liver damage), and references are cited showing this. However, one would think that Dr. Al-Bayati would know that steatosis (a fatty infiltrate of the liver cells) is not the same thing as any of these. Moreover, the pathology report did not state that Eliza Jane's liver had any of the more common manifestations of amoxicillin-clavulanate-induced liver injury, casting further doubt on the antibiotic as the cause. (Drugs more classically associated with steatosis include valproic acid, tetracycline, amiodarone, and aspirin, not amoxicillin-clavulanate. Indeed, a PubMed search on "steatosis" and "amoxicillin" failed to turn up a single reference.) I didn't see any of Dr. Al-Bayati's references supporting his implication that amoxicillin could cause steatosis within a day of exposure. Clearly, the steatosis must have been a pre-existing condition. As for the pulmonary edema and ascites reported, that could be due to allergic reaction, septic shock from an ear infection, cardiopulmonary collapse from whatever cause, or a number of other factors, but in the context of Eliza Jane's HIV infection, rapid deterioriation, and cardiovascular collapse, it fits.

The bottom line is that you can compare these two sets of conclusions about what caused Eliza Jane's death and decide which seems more plausible. The first, the coroner's report, looks at a child of an HIV-positive mother who refused to take AZT during pregnancy, breast-fed her child even though that is known to increase the rate of transmission of HIV, and refused to have her child tested for HIV, a child who collapsed after upper respiratory and ear infections, whose autopsy findings showed HIV encephalitis, P. carinii in the lungs, and severe anemia. Given such findings, it's hard not to conclude that Eliza Jane died of AIDS complications. (Whether or not Dean's contention that the coroner had some sort of horrendous bias or was grandstanding is true, I cannot say, but certainly Dean hasn't provided any compelling evidence that he was.) The second hypothesis, promulgated by an HIV "dissident" with a definite axe to grind, requires us to believe in not just one, but three, highly unlikely occurrences (plus one simply unlikely occurence), namely:

1. A previously completely healthy girl developed PVB19 infection leading to both encephalitis and aplastic anemia (possible, but highly unlikely, and, even if PVB19 were found, it would be far more likely that it was able to cause anemia because of immunosuppression due to AIDS).
2. This same girl also developed an acute allergic reaction to amoxicillin that led to cardiovascular collapse and--oh, by the way--also caused steatosis of the liver within a day after starting the drug, the steatosis being something even Dr. Al-Bayati's own references do not seem to support as being likely.
   
3. This same otherwise healthy girl had sufficient quantity of P. carinii in her lungs to show up on Gomori methenamine silver staining at her autopsy.
4. The medical examiner and neuropathologist either botched the staining for the p24 protein (or that it was a false positive) and an experienced neuropathologist didn't know the pitfalls of the diagnosis of HIV encephalitis using brain tissue sections.
   

A veritable herd of zebras indeed. To believe Dr. Al-Bayati's scenario, you have to believe that at least four very unlikely things happened in the same case, rather than the very likely conclusion (based on the autopsy findings) that AIDS killed Eliza Jane.

Not surprisingly, credulous guy that he is when it comes to anything that supports HIV denialism, Dean does exactly what he accuses Richard Bennett of and "sucks down any codswollop he's fed on this subject and spews it back out on command." No doubt Dean will likely accuse me of the same and/or consider my response to him to be "dishonest" or possibly lacking class. We'll see. Part of the reason I haven't addressed Dean's "skepticism" about AIDS much before is because observing him pontificate on AIDS has taught me that arguing with him on this issue is completely pointless. Quite frankly, even in this instance, it's unlikely that I would have bothered to respond if Dean hadn't annoyed me so much by impugning my honesty.

In the meantime, I plan on submitting this to Grand Rounds this week. (That's about as close to "peer review" as you can get in the blogosphere.) I'm also very interested in what other doctors, scientists, and medbloggers have to say about this. I'm not an AIDS expert, but a lot of the stuff in Dr. Al-Bayati's report is so off the wall that it didn't take much to find the inconsistencies. (And if I, a knowledgeable non-expert, can pick the flaws apart, imagine what a real expert could do.) Nonetheless, if I got something grievously wrong that calls into question my analysis, I'd like my peers to let me know.

ADDENDUM: An update including the opinions of two additional physicians has been posted here. They cover a lot of the same ground, but from different viewpoints and different areas of expertise and, when combined, present a truly devastating rebuttal to Dr. Al-Bayati's "report."