But Why Stop Now? Urinary Porphyrin Data of Geier et al Should Have Been Controlled for Age

In their recently published study in the Journal of the Neurological Sciences, Geier et al not only presented values for transsulfuration metabolites in children with autism spectrum disorder (ASD); they also measured urinary porphyrins. Their rationale for doing so was to test a pet theory that autism is due to a reduced ability to excrete environmental mercury, as a result of an innate "decreased detoxification capacity"—which is proposed to be characterized by altered levels of transsulfuration metabolites. (The questionable values for some transsulfuration metabolites presented by Geier et al in the JNS article and elsewhere have been discussed here, here, here, and here.) In addition, because investigators have not been able to consistently find elevated mercury levels in children with ASD by direct measurement, Geier et al have turned to the use of "mercury intoxication-associated urinary porphyrins" as markers of mercury toxicity (most notably from thimerosal-containing vaccines) in children with ASD. In the JNS article, they write,

[I]t was previously demonstrated that the transsulfuration pathyway products of glutathione [15] and sulfate [16] were related to mercury excretion rates, and that the heme synthesis pathway products of urinary porphyrins can provide specific profiles that reflect mercury toxicity [17].

The references cited by Geier et al in this statement are worth examining.

First, reference "15" is a 1985 article from Ballatori and Clarkson, with the relatively straightforward (if not bone-dry) title, "Biliary secretion of glutathione and of glutathione-metal complexes." This study is an examination of the excretion of methylmercury in the bile in rats, which was found by the authors to closely parallel the biliary secretion of reduced glutathione (GSH). Important pharmacokinetic differences between methylmercury and ethylmercury (which is in the vaccine preservative thimerosal) have been discussed at this blog and by many others. Also the biochemical/biophysiologic leap from rodents to humans should have been acknowledged by Geier et al.

The same criticism can be applied to the use of reference "16," a 2004 review of organ systems in the American lobster (!) that regulate and detoxify environmental heavy metals.

Reference "17" is a 1996 study by James S. Woods from the University of Washington. In rats, Woods demonstrated changes in urinary porphyrins after prolonged exposure to—again—methylmercury. Specifically levels of 4- and 5-carboxyl porphyrins and the expression of precoproporphyrin were demonstrated in the exposed animals. Woods also conducted a possibly controversial pilot study in Mexican dental workers who were exposed to mercury-containing dental amalgam. Woods showed that the excretion of mercury increased and that levels of these urinary porphyrins decreased with chelation.*

In the JNS study, Geier et al measured urinary porphyrins in their 28 subjects with ASD (age range, 2-16 years), but not in their control children—as they did when assessing transsulfuration metabolites. (The reason for not measuring urinary porphyrins in the control group is not explained by the authors.) So without a true control group, Geier et al compared urinary porphyrin values** in their 14 children with "mild" ASD (CARS score ≤38.5) with those in 14 children with "severe" ASD (CARS score ≥38.5). (It's not clear what the authors did with the kids who hit the 38.5 mark.)

Not too surprising, Geier et al claimed significant differences in urinary levels of pentacarboxyporphyin (ie, 5-carboxyl porphyrins) and precoproporphyrin (Table 2) between the mild and severe ASD groups, which would be consistent with the methylmercury rat data of Woods. The authors also ostensibly monkeyed around with the various ratios of urinary porphyrins and found other significant differences between the 2 ASD groups. Additional fiddling demonstrated relationships between the CARS score and some porphyrin ratios. These data are intended to show that the authors' surrogate markers for mercury intoxication—urinary pentacarboxyporphyin and precoproporphyrin—are associated with the severity of autism.

Last, Geier et al assessed the plasma oxidized glutathione (GSSG) levels—as a "strong indicator of cellular oxidative stress"—among ASD children with low urinary porphyrins or high urinary porphyrins and claimed significantly increased levels of plasma GSSG in subjects with high urinary pentacarboxyporphyin or precoproporphyrin levels. Again, Geier et al intended to demonstrate that their surrogate markers for mercury intoxication are associated with a reduced capacity to excrete mercury, per the GSSG level, in ASD children. The main problem with this particular finding is that the plasma GSSG values presented by Geier et al are considerably different (eg, by 3 orders of magnitude) from those published elsewhere, including references cited by the authors.

Also published data suggest that Geier et al should have controlled for age-related differences in urinary porphyrin excretion—especially given that their subjects ranged in age from 2-16 years. A 1996 study by Minder and Schneider-Yin (Age-dependent reference values of urinary porphyrins in children) found distinctive age-related changes in the urinary excretion of 3 porphyrins, which may be explained—depending on the porphyrin—by age-related changes in the physiologic development of the excretion system and heme synthesis. For instance, their data showed that the urinary excretion of coproporphyrin III decreases steadily from the age of approximately 2 years to late adolescence. In an e-mail response, lead author Elisabeth Minder stated, in reference to the JNS study by Geier et al, that "one should control the data for age."

CARS = Childhood Autism Rating Scale. 

* Curiously enough, Woods is a coauthor of a 2006 JAMA article, which reported no significant differences in urinary mercury levels or neurologic function between Portuguese children who received dental amalgam and those who received a resin-based composite for routine dental work. The study's ethics were criticized in the Petition to Order Mercury Amalgam Withdrawn From Interstate Commerce. According to the JAMA article, "Urinary...porphyrins were monitored as indicators of renal responses to mercury..and will be reported separately."

** Urinary porphyrin levels from the subjects, who were recruited in Dallas, Texas, were shipped to and measured at, for some inexplicable reason, the Laboratoire Philipe Auguste in Paris. A coauthor of the JNS article is Robert Nataf from the same Paris lab. Nataf is the lead author of a retrospective 2006 study, which reported relatively elevated coproporphyrin and precoproporphyrin levels in children with autism.