Franšais      print      e-mail    


Ambroise D., Blech M.F., Di Majo P. and Ph. Hartemann (1995) Assessment of the Health Risk Associated with THM in Dialysis Water. Rev. Sci. Eau 8 (2) 261-275. [article in French]

Original title: Évaluation du risque toxique lié à la présence de trihalométhanes dans l'eau utilisée pour la dialyse.

Full text (PDF)


Hemodialysis is an indispensable therapy for patients with chronic renal failure. Two or three times a week and over several years, their blood is dialyzed in an artificial kidney against a dialysis fluid called dialysate.

Each time, 90 to 200 liters of this fluid will flow through the apparatus. Before being mixed with the dialysis concentrate, the water will be treated in order to eliminate harmful substances such as aluminum or endotoxins.

Trihalomethanes (THM) are probably the most widespread chlorination byproducts of tap water. Most of them are known as carcinogens for animals and mutagens in vitro. Although their hepatotoxicity and nephrotoxicity are obvious after acute intoxication, their effects at low doses on human health have still not been clearly demonstrated.

Considering the great amount of water required by hemodialysis patients, we found interested in determining wether the control of these substances by the hospital water treatment plant was efficient. We decided then to analyze weekly and during two months, the tap water of two hemodialysis departments for THM, before and after various forms of treatment. The treatment in both departments was the same and made up of four important stages: chlorination, softening, charcoal filtering and reverse osmosis.

THM determinations were conducted using the headspace technique with a gas chromatograph equipped with a split injector and an [sup]63Ni electron capture detector.

Our results show that chloroform and dichlorobromomethane were present in tap water. Their respective mean concentration in both department came to 56 µg/l and 5 µg/l. After chlorination and water softening, these figures had moderately but significantly increased. In the first department, thanks to new granular activated carbon, a large part of THM (especially dichlorobromomethane) had been removed. However after seven weeks, this treatment was no longer efficient and only 7% of the influent chloroform and 50% of the dichlorobromomethane could be removed. In the second department, the charcoal filter had already been working for more than one year at the beginning of our study. No decrease of the chloroform concentration had been observed and dichlorobromomethane had significantly increased. 80 to 90% of influent THM were removed after the double stage of reverse osmosis using polyamide membranes.

With new granular activated carbon, the dialysis fluid only contains 1 µg/l of chloroform. But after seven weeks or more, it will reach an average of 10.5 g/l of chloroform and 1 µg/l of dichlorobromomethane. These figures are probably underestimated as our study was performed in winter and THM concentrations are less important during that season.

These results mean that during a single session, 0.9 to 2.1 mg of chloroform will reach the artificial kidney. Depending on the weight of the patients, this exposure will be equivalent up to 10 times the value recommended by the World Health Organization (WHO) for drinking water.

The last part of our study monitored the chloroform concentration in dialysate coming out the artificial kidney during an hemodialysis period. A significant decrease, reaching up to 45% of the influent amount, was observed. This result suggests that some of the chloroform must have crossed the dialysis membrane.

According to all these results, we think that it would be of great interest to explore the metabolism of chloroform on hemodialysis patients and to search for eventual toxic effects. Practical advices to people in charge of water treatment plants in hemodialysis department would be to use raw water with low concentrations of humic materials, in order to restrict THM formation. The charcoal filter should be changed more often (probably after 6 or 7 weeks). Alternatively, ways could be found for rapid regeneration of charcoal for THM removal. Finally and according to previous studies, a polyamide membrane should be systematically used for reverse osmosis.

Our study could eventually be completed by searching in the dialysis fluid any other chlorination by-products which are responsible to a large extent for tap water mutagenicity.


Trihalomethanes (THM), chloroform, hemodialysis, health risk.

Corresponding author

Ph Hartemann, Laboratoire d'Hygiène et de Recherche en Santé Publique, 11bis rue Gabriel Péri, BP 288, 54515 Vandoeuvre-lès-Nancy, FRANCE

Telephone : +(33) / Fax : +(33)

Franšais      print      e-mail    

Update: 2006-12-19
© INRS Eau, Terre et Environnement