Evaluation of a report of effects of propylene glycol on keel-billed toucans  by Lynn Pottenger, PhD (DOW Chemical)

A publication, by Worell et al. (2000)[1] describes several cases of keel-billed toucans in zoos becoming ill, with some birds that did not respond to supportive treatment and died.  For several of the cases, the publication lists “before” and “after” diets administered to the birds, and for one case report, the publication describes administration of a medication to an already sick bird.  The publication asserts that the clinical effects, and subsequent deaths of some birds, were due to propylene glycol, present in the “after” diet or in the medication, respectively. 

In order to investigate the likelihood that propylene glycol was involved in the clinical effects and eventual deaths described, we obtained samples of the diets mentioned in the publication and subjected them to chemical analysis for propylene glycol content.

 Table 1 lists the commercial names of the diets investigated and the propylene glycol content quantified in each, using a standardized analytical methodology (GC/FID detection).  The limit of detection was 500 ppm, or 0.05% (w/w).

Table 1. PG content in brandname animal diets cited in Worrell et al. (2000)[2]

As can be seen from Table 1, only one diet contained quantifiable levels of propylene glycol.  Thus one of the “after” diets, which were described as causing the deaths of keel-billed toucans based on their content of propylene glycol, did not have any quantifiable amount of propylene glycol.  The other “after” diet did contain propylene glycol, at 2.5% (w/w), a level which has been shown to cause no effects in repeated exposures to quails and chicks[9]^,[10], in addition to many mammalian species (rats, mice, rabbits, dogs, guinea pigs, etc.).

Based on these analytical determinations of propylene glycol levels present in the commercially available diets cited in Worrell et al. (2000), it is highly unlikely that propylene glycol was the cause of any of the clinical effects, including deaths, described in the keel-billed toucans, and attributed by Worrell et al. (2000) to administration of the “after” diets.  It is not clear what may have been the causative factor in these incidents.

This conclusion is supported by the published literature demonstrating no effects from feeding quails and chicks diet including up to 2.5% PG (w/w).  In fact, according to the USEPA and OECD, PG is an accepted oral carrier for guideline toxicity testing in birds.

An additional critical flaw in the Worrell et al. (2000) publication is the lack of any controls for the hypothesis of PG toxicity.  There were no healthy birds fed defined PG-supplemented diet, nor any clearly identifiable birds that received defined PG-free diets, nor any birds that did not change diet.  With the limited information available, there is no way to rule out the possibility that the clinical effects reported are unrelated to the low iron diets.

In conclusion, the Worrell et al. (2000) publication does not offer any proof that PG was related to the clinical effects, including deaths, reported by the authors.  The analytical results on PG content of the diets conducted by Dow and reported here, do not support any causative role for PG in the clinical effects reported in keel-billed toucans by Worrell et al. (2000).