Health Stream Literature Summary - Issue 58 - June 2010

Human health risk assessment of carbamazepine in surface waters of North America and Europe.
Cunningham VL, Perino C, D'Aco VJ, Hartmann A and Bechter R. (2010) Regulatory Toxicology and Pharmacology, 56(3); 343-351.

Carbamazepine (CBZ) is a therapeutic pharmaceutical agent for the treatment of epilepsy, trigeminal neuralgia, bipolar depression, excited psychosis and mania. Since 1998, CBZ has been detected in sewage treatment plant (STP) influents, STP effluents, surface waters, drinking waters and groundwaters in Europe, the United States and Canada. CBZ's greater frequency of detection may be because it appears to be more resistant to removal than other pharmaceuticals and is used in relatively high volumes. There is a considerable body of literature on its metabolism and pharmacology. A human health risk assessment was undertaken for environmental exposure to CBZ and two of its major metabolites, carbamazepine diol (CBZ-DioH) and carbamazepine N-glucuronide (CZB-N-Glu) in the aqueous environment.

CBZ itself is the pharmaceutically active entity and of all its metabolites, only 10,11-epoxide (CBZ-EP) (comprising 1-2% of the excreted dose) can significantly add to the antiepileptic effect of CBZ in people. The major metabolite CBZ-DioH is inactive. There is no pharmacological data on CBZ-N-Glu, however as a highly polar metabolite it would not be expected to demonstrate any therapeutic activity. CBZ, CBZ-DiOH, and CBZ-N-Glu, which make up approximately 64% of the excreted dose, were the compounds selected for assessment. The Acceptable Daily Intake (ADI) represents a level of daily intake that should not result in an adverse human health effect from direct exposure in a population, including individuals which are particularly sensitive. The ADIs for CBZ, CBZ-DiOH, and CBZ-N-Glu were estimated. The ADIs for CBZ-DiOH, and CBZ-N-Glu were set at 333 micro g/kg/day. The ADI for CBZ was 15.9 micro g/kg/day. The ADIs were combined with standard assumptions about potential exposure, via drinking water and fish consumption to derive predicted no effect levels (PNECs) for CBZ, CBZ-DiOH, and CBZ-N-Glu. Three different types of PNECs were derived for both children and adults: a PNEC protective of drinking water exposures only (PNECDW); a PHEC protective of fish consumption exposures only (PNECF); and a PNEC protective of combined drinking water and fish consumption exposures (PNECDW+F). The PNECs for children are more conservative for screening purposes than for adults as children are assumed to drinking more water and eat more fish per body weight than adults. Only the PNECs calculated for children were therefore presented. Combined drinking water and fish consumption human health PNECs for CBZ, CBZ-DiOH, and CBZ-N-Glu in children were 226,000, 4,760,000 and 4,760,000 ng/L respectively.

Measured environmental concentrations (MECs) for CBZ in surface waters of North America and Europe were calculated. There were 2046 surface water MECs for CBZ from 52 references included in the analysis. Concentrations of CBZ reported for European surface waters range from non-detectable to 11,581 ng/L, while concentrations for Canadian and U.S. surface waters range from non-detectable to 1,500 ng/L. CBZ-DioH has been reported for one sample in Canadian surface waters at a concentration of 2.2 ng/L and from three rivers in Germany at a maximum concentration of 140 ng/L. The presence of CBZ-N-Glu has not been reported in surface water. CBZ in finished drinking water or drinking water source water has been found at levels from less than 0.8 to 258 ng/L .

Predicted environmental concentrations (PECs) for CBZ, CBZ-DiOH, and CBZ-N-Glu in North American and European surface waters were estimated using the PhATE and GREAT-ER models, respectively. The PECs estimated using PhATE were compared with MECs reported for CBZ in North American surface waters. The cumulative probability distributions of all MECs (n=464) and of PECs for all model segments (n=2710) at both low and mean flow conditions were calculated. The low-flow PECs from PhATE and GREAT-ER models were considered to be conservative estimates of exposure as compared with measured concentrations.

In order to evaluate the potential human health risk associated with exposure CBZ, CBZ-DiOH, and CBZ-N-Glu, combined drinking water and fish consumption PNECs were compared with MECs and PECs. The results of these comparisons are expressed as margins of safety (MOS) or the ratio of the PNEC to the MEC or PEC. If the MOS is large (MOS greater than or equal to 1) then the measure or predicted concentrations are far below allowable concentrations. The MOS for CBZ, CBZ-DiOH, and CBZ-N-Glu in both North America and Europe were found to be high, ranging from 340 to 6560. These MOS indicate that there is no appreciable risk to human health from environmental exposures from drinking water and fish consumption. This analysis represents conservative assumptions and data. The MECs and PECs evaluated represented only surface waters, not drinking waters. For MECs in particular, many of the samples analysed came from areas expected to be impacted by sewage effluents, not from areas used for drinking water diversion. If the approach was refined and drinking water concentrations were taken into account, the MOS would probably increase. This assessment does not however consider the potential effects of mixtures.

© Copyright Water Quality Research Australia Limited http://www.wqra.com.au/
Health Stream articles may be reproduced and communicated to third parties provided WQRA is acknowledged as the source. Literature summaries are derived in part from copyright material by a range of publishers. Original sources should be consulted and acknowledged.