Dibutyl phthalate

Exposure related observations in humans: other data

Administrative data

Endpoint:

exposure-related observations in humans: other data

Adequacy of study:

other information

Data source

Reference

Reference Type:

publication

Title:

Unnamed

Year:

2013

Applicant's summary and conclusion

Executive summary:

Extractability of phthalate plasticizers

Extractability of phthalates from plastic articles as a function of composition, weight, surface area and time (migration rate) has been studied in vitro by a number of groups using various mechanical methods including shaking, ultrasound, tumbling (‘head over heels’) and impaction (Babich, 2002). Studies using these different methods have generated a broad range of results depending on the experimental conditions.

In vivo, phthalate extractability has been studied using adult volunteers providing saliva samples during mastication of plastic articles to measure migration of the plasticizer into the saliva as a function of time (migration rate).

These studies allow a direct comparison of results from in vivo and in vitro mechanical methods. In the majority of the studies, results from the in vitro methods underestimate the migration of phthalates from chewed articles. The results for in vitro studies were therefore not considered to be as useful as those from in vivo studies in determining suitable migration rates for calculating systemic doses.

DINP is the most prevalent phthalate in children’s toys and the migration of this chemical from plastics has been studied most extensively. The studies demonstrate that migration of phthalates from plastic products is determined more by the magnitude of mechanical action applied to the plastic rather than the chemical diffusive properties determined by the physicochemical characteristics of the substrate or concentration of phthalate.

Chen (1998) conducted an in vivo study in the US with adult volunteers and an in vitro study using impaction methods and saliva simulants. In the in vivo study, two plastic disks (each with a surface area of approximately 10.3 cm2) were cut from each of five identical PVC toy ducks, each containing 43 % DINP by weight. Ten US Consumer Product Safety Commission (CPSC) staff volunteers were asked to gently chew the disks for four 15-minute intervals. Saliva samples were collected after each chewing interval and analysed for DINP. Migration rates varied substantially from individual to individual. The average DINP migration rate across all time periods from volunteers was 26.03 μg/cm2/h (range 6.14–57.93 μg/cm2/h). In vivo migration rates also averaged 39.5 times higher than rates obtained from the in vitro impaction study. In vitro impaction studies of phthalate release rates (range 0.1–4.4 μg/cm2/h) from samples of children’s toys or child-care products showed poor correlation between release rates and the amount of phthalate present in samples.

Meuling and Rijk (1998) conducted an in vivo study in the Netherlands with 20 adult volunteers and an in vitro study with a simulant of saliva using shaking, head over heels mixing and ultrasound methods. In the in vivo study, three specimens were used: a standard PVC disk (38.5 % DINP), part of a PVC teething ring (43 % DINP), and a disk punched from the same teething ring (43 % DINP). Each specimen had a surface area of 10cm2. Initially all 20 volunteers were asked to suck and bite on the standard PVC disc for four 15-minute intervals. Saliva samples were collected after each biting interval and analysed for DINP. Subsequently, the volunteers were divided into two groups of 10. One group repeated the test using part of the teething ring while the other group used the disk punched from the teething ring. In the in vivo study, the mean release rates were: 8.28 μg/cm2/h (range 1.8–49.8 μg/cm2/h) for the standard PVC disc, 14.64 μg/cm2/h (range 5.4–53.4 μg/cm2/h) for the teething ring and 9.78 μg/cm2/h (range 5.4–34.2 μg/cm2/h) for the disc punched from the teething ring. The researchers noted that the amount of DINP released into saliva exceeded its expected solubility and that mechanical force was required in the in vitro studies in order to attain migration rates comparable to that obtained from the in vivo studies.

Fiala et al. (1998) conducted an in vivo study in Austria with nine volunteers and an in vitro study with a simulant of saliva using shaking or ultrasound methods. In the in vivo study, PVC sheets (32 % DEHP) and parts of PVC teethers (36 % DINP) were used separately. Each specimen had a surface area of 10–15 cm2. The volunteers were asked to suck only or chew the samples separately for 1–3 hours. Saliva samples were collected and analysed. For DINP, the mean release rate (sucking for one hour) was 8.33 μg/cm2/h (range 2.97–14.52 μg/cm2/h). Higher values were recorded from chewing. The mean release rate for DINP (chewing for one hour) was 13.3 μg/cm2/h (range 7.68–21.52 μg/cm2/h). This study also showed that migration rates were substantially higher in the in vivo chewing study than those obtained in the in vitro studies.

Niino et al. (2001) conducted an in vivo study in Japan with four volunteers and an in vitro study with a simulant of saliva using shaking methods. In the in vivo study, two PVC ball samples were used: sample A contained 10.0 % DBP and 18.5 % DEHP, and sample B contained 25.6 % DINP. Each specimen had a surface area of approximately 15 cm2. Four volunteers were asked to gently chew each of the specimens for four 15-minute intervals. Saliva samples were collected after each chewing interval and analysed for phthalate content. In contrast to previous studies, the in vitro study of phthalate migration showed a substantially higher mean migration rate at approximately two orders of magnitude higher than the human in vivo study.

In a follow-up study, Niino et al. (2002) conducted an in vivo study with four volunteers and an in vitro study with a simulant of saliva using shaking methods. In the in vivo study, samples of a PVC plate and toys (including pacifier, teether, rattle, ball, soft doll, containing 16.0–58.3 % DINP) were tested separately. Each specimen had a surface area of approximately 15 cm2. Four volunteers were asked to chew each of the specimens for four 15-minute intervals. Saliva samples were collected after each chewing interval and analysed for DINP. The average migration rate across all samples was 16.4 μg/cm2/h (SD 2.8 μg/cm2/h). The highest migration rate was for the PVC plate sample at 32.6 μg/cm2/h (SD 2.6 μg/cm2/h). The authors noted that DINP contents in the toy products did not correlate with the amount of in vivo migration. The in vitro migration studies showed consistently higher mean migration rates than the in vivo studies.

Selection of migration rate for exposure assessment

As the results from the in vitro studies do not reproduce the in vivo findings for the same systems, the results from only in vivo studies are used in the exposure assessment. The following conclusions can be drawn from the above five in vivo studies:

• Within studies, migration rates vary substantially from individual to individual, even though the same action (e.g. chewing) is involved.

• Migration rates have little direct relationship with the phthalate content of an article in the tested phthalate range of 15–58 % by weight, indicating that differences seen between test articles may depend more on the properties of the PVC grade comprising the article.

• The amount of phthalate released into saliva through biting and chewing exceeded its expected solubility in water in all in vivo studies, indicating that migration is not merely a simple diffusion process.

• Migration rates are proportional to the amplitude of mechanical action i.e. chewing results in a higher migration rate than mouthing or sucking alone.

Based on the above conclusions, it is evident that migration of phthalate plasticisers from plastic toys into saliva through biting and chewing is the combined effect of molecular diffusion and mechanical action, with the latter likely to be the dominating factor. The migration rate of phthalates from articles appears largely determined by the magnitude of the mechanical force applied to an article, and the properties of the PVC grade comprising the article, and less affected by the physicochemical characteristics or concentration of a particular phthalate.

The migration rates determined for DINP under chewing condition can be extrapolated to other phthalates assuming similar product uses and concentrations in products.

In these studies, the use of adults in in vivo studies as a surrogate for the activities of children is accompanied by several uncertainties. Firstly, the level of mechanical force applied to the plastic toys may differ. Therefore, the use of adults in the in vivo studies might lead to an overestimation of phthalate migration from toys. Also, children do not swallow all the saliva, which means that estimates of exposure from adult in vivo studies, where all saliva harvested is assumed to be swallowed, may again overestimate the oral exposure of children. Finally, absorption through the oral mucosa is not accounted for in migration measurements in adults in vivo. However, compared to potential oral ingestion, mucosal absorption is likely to be very low.

The highest in vivo migration rate observed for DINP in a well-conducted study was 57.93 μg/cm2/h from articles with up to 54 % DINP content (Chen 1998). This migration rate is therefore applicable for a worst-case exposure assessment for children from the use of DINP in toys. The mean migration rate for DINP in this study was 26.03 μg/cm2/h (Chen 1998), which is similar to the highest mean migration rate of 32.6 μg/cm2/h (Niino 2002) in a study using a smaller number of volunteers. The mean migration rate determined by Chen (1998) is regarded as applicable for typical exposure assessment in toys.