Registration Dossier

Ecotoxicological information

Long-term toxicity to fish

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

In summary, there is no reliable long-term study indicating adverse effects to fish at or below the “apparent” water solubility of DEHP or when administrated via food. Therefore, it is not considered suitable to specify a chronic NOEC for fish exposed via water or food.

Key value for chemical safety assessment

Additional information

As reported in RAR, 2008, most chronic toxicity studies available are not reliable for assessment. The two main reasons are that test concentrations were all above water solubility of DEHP (exposure via water) and/or that effects observed cannot be firmly attributed to DEHP(exposure via water or food).

Today no new data are available. The studies have been re-evaluated for the purpose of REACH regulation and chemical safety assessment and have highlighted one key study. Considering the very low solubility of the substance and its adsorption behaviour, the study from Caunter et al. (2004) appeared to be the most relevant study in terms of worst condition that could be encountered in the environment. Authors studied the effects of DEHP on three generations of Pimephales promelas. Animals were exposed in a flow through system via the water (nominal 5 µg/L constant, measured 3.2 µg/L and 2.6 µg/L in low dose and high dose food group, respectively) and food supplemented at a low dose level (nominal 125 mg/kg, measured 115 mg/kg) or at a high dose level (nominal 500 mg/kg, measured 474 mg/kg). Solvent and water controls were performed and the parameters followed were: Hatching success, growth, egg production, survival, concentrations in vitellogenin and sex ratio.

A number of statistically significant differences between treatments and controls were observed in the study. In most cases however, these differences were probably not caused by the exposure of DEHP. Due to the large variation, frequent statistically significant differences between dilution water and solvent control groups, inconsistent and often not dose related responses in the DEHP treated groups and for some endpoints a small group size, it is hard to draw firm conclusions regarding the effects seen. It is therefore not possible to derive a NOEC from the study and thus it is not possible to use the results from the study in a quantitative risk characterization. However, the exposure levels via food were fairly high, much higher than normally seen in the environment. And since no serious effects that can be unequivocally attributed to the exposure of DEHP were observed, the study gives a fair indication that serious effects probably should not be anticipated at more environmentally realistic exposure levels. 

The studies reported in the RAR showing effects of DEHP on sex ratio at nominal concentration of 16500 and 800 mg/kg (Norrgren et al.,1999 and Norman et al., 2007 respectively) were not considered as relevant in the present assessment. As a matter of fact, Caunter et al. 2004 have taken into account the previous studies and tested concentrations in food at lower, albeit still high concentrations compared to environmentally relevant levels: No serious adverse effects were attributable to DEHP at simultaneous exposure via water and food (three generation study, see above). Thus we assume that at environmentally realistic exposure levels no effects on fish are to be anticipated.

Two supporting studies corroborate this conclusion:

In DeFoe et al. (1990), Rainbow trout eggs were exposed for 90 days up to 0.502 mg/L and no statistically significant effects on hatchability, survival or wet weights were observed in any of the test concentrations (ranging from 48 to 502 µg/L). The statistically not significant reduction of mean wet weights by 9.7 and 10.2% compared to control weights at the two highest exposure concentrations (259 and 502 µg/l) might nonetheless be attributable to DEHP. However, taking into account the height of test concentrations being far above the true water solubility of approximately 3 µg/L, this slight reduction in growth may well be due to physical effects caused by oversaturation.

Metcalfe et al. (2001) assessed estrogenic properties of DEHP in vivo up to 5000 µg/L using Japanese medaka and in vitro up to 100 mg/L in a yeast estrogen screening (YES) assay. No effects on sex ratios, incidence of intersex or morphometric parameters could be observed in the medaka and no estrogenic activity could be detected in the YES assay.

Hence, even with DEHP exposure concentrations well above water solubility, no serious adverse effects on fish could be observed. Slight observed effects were either inconsistent and not clearly attributable to DEHP or most probably due to physical effects at high exposure concentrations exceeding water solubility by far.

Recently, a new study was published (Zanotelli et al., 2010) investigating long-term effects of DEHP on growth (size and weight differences assessed) of Guppi (Poecilia reticulata) fish larvae (less than 1 week of age) at 28°C for 91 days. DEHP concentrations of 0.1, 1.0 and 10 µg/L were applied. From the published data a NOEC (91 days, length and weight) of 0.1 µg/L and a LOEC (91 days, length and weight) of 1 µg/L can be derived. However, several severe draw-backs of the study (beyond others: no analytical monitoring, no measurements of dissolved oxygen concentration, no data on the origin of fish and especially neither weight distribution nor length distribution of fish at the start of the test) put obtained results into question and the study is regarded to be not reliable.

In summary, there is no reliable long-term study indicating adverse effects at or below the “apparent” water solubility of DEHP or when administrated via food. Therefore, it is not considered suitable to specify a chronic NOEC for fish exposed via water or food.