Health Stream Article - Issue 29 March 2003

Copper Toxicity in German Infants

A study of potential copper toxicity in infants has shown no indications of liver malfunction from exposure to copper levels in municipal tap water in Berlin. The study involved the testing of water samples from households with children up to 18 months of age, followed by a medical examination of a subgroup of children considered to have high copper exposure. The investigation was prompted by public concerns in Germany over a number of cases of severe liver disease, cirrhosis and some fatalities in children that were reportedly linked to high copper levels in tap water.

Families with infants were identified from the federal birth registry, and information about the study was mailed to a total of 12,000 households across Berlin. Families who were willing to participate were asked to complete and return a consent form. Each household was then sent two sampling bottles with instructions on how to collect composite water samples. The first sample type was made up by collecting 100ml aliquots from each tap in the house every time it was used during a day. The second sample type was taken the following day by collecting 250ml aliquots at four times during the day (morning, noon, evening and before going to bed).

Families were also asked to complete a questionnaire on the age of the dwelling, pipe materials, any construction or renovations that had been undertaken, the water intake of children, and the duration of feeding with breastmilk and baby formula made with tap water. A total of 4,404 households returned consent forms and 59% of these (2619 households) sent the two composite water samples for analysis as requested. An additional 325 households took part in a preliminary study with different sampling methods.

Overall, 804 households (29.9%) were found to have copper levels of 0.8mg/L or higher in one or both of the composite samples. Comparison of mean copper levels with the reported age of the water pipes in households (regardless of pipe material), showed a decline in copper concentration until the 35-40 year age bracket, followed by a rise in concentration in older houses. When only households stated by the householder to have copper pipes were included in the analysis, a similar relationship was seen although the minimum copper levels occurred in houses that were 40-45 years old.

The results of copper analyses on the samples from the 2,619 households in the main study are summarised in the table below (units = mg/L):

In order to select a subgroup of infants with relatively high copper exposure from tap water, the following criteria were used:

• copper concentration of 0.8 mg/L to less than 2.0 mg/L in one or both composite water samples, and
• infant with tap water intake of 200ml or more per day for at least 6 weeks
  or
• copper concentration of 2.0 mg/L or higher in one or both composite water samples, with any reported tap water intake

The value of 0.8 mg/L was chosen as it is slightly above the reported average concentration for copper in human breast milk. Families with an infant in the high exposure group (541 infants) were recommended to have a paediatric examination performed on the child, including a blood analysis and examination of the liver by ultrasound and manual palpation.

A total of 517 children underwent physical examination but blood samples were collected from only 183. No cases of suspected liver damage were reported by the examining paediatricians on the basis of reported health status, symptoms and liver palpation. Blood samples were tested for markers of copper metabolism and liver function, including serum copper, ceruloplasmin, total bilirubin and several liver transferase enzymes. Immunoglobulins and C-reactive protein were also measured as indicators of infections which may have affected other parameters. The duration of tap water exposure in the 183 children with blood samples was variable, with 34 receiving at least some drinking water from birth, including 8 who had never been breast fed.

Of the 183 children with blood samples, 2 were excluded from analysis due to evidence of a poor quality sample (blood cell lysis) and a confirmed infection with Toxoplasma gondii. Among the remaining 181 children, 8 were found to have outlying values for liver enzymes or serum copper levels (defined as more than 4 standard deviations from the mean value, or with a probability of chance occurrence less than p=0.05). Six of these had diagnosed infections and one had a liver haemangioma (a benign tumour consisting of dilated blood vessels) and uretic obstruction. Five children had slightly unusual ultrasound imaging of the liver or spleen, and two of these had known infections at the time of blood sampling. None of these observations was considered to be of clinical significance. The researchers noted that since examination or blood sampling from children in the low exposure range was not carried out due to ethical considerations, there was no concurrent control group for comparison. Reference values for blood parameters in normal infants were taken from literature sources, but such data are limited and often based on small numbers of individuals.

A number of statistical tests were performed to determine whether there was any significant relationship between estimated daily and total copper intakes from water and the tested blood parameters. The children were divided into three groups in the basis of copper intake, and no significant differences were found between the groups for liver transferase enzymes, serum copper, total bilirubin, immunoglobulins, ceruloplasma or free copper. The results were not affected by inclusion or exclusion of the outlier values. Subgroup analysis was also performed for children who had been examined within 14 or 21 days of their parents receiving letter of the recommendation for a medical examination. This was done to minimise the effect of any change in copper exposure that may have occurred if parents avoided or reduced tap water exposure of the child after notification. Again, no significant relationship was found between copper intake and the measured blood parameters.

The researchers concluded that these observations indicate that exposure to the copper levels found in municipal tap water in Germany is not associated with liver malfunction or detectable changes in a number of biochemical markers of liver function.

In discussing the previously reported cases of copper-related liver disease in children in Germany, they note that the majority of these cases were identified retrospectively from medical records and consequently the copper exposure levels are not well documented. The households involved were all supplied by private wells, and it was suspected that excessive copper exposure had occurred due to leaching of copper from household pipes by the acidic groundwater. Some of the affected families were related to each other, suggesting the possible involvement of an inherited predisposition although no specific gene defect(s) have yet been identified.

Reference: Epidemiological investigation on chronic copper toxicity to children exposed via a public drinking water supply
Zietz BP, Deiter HH et al. (2003) The Science of the Total Environment 302 127-144.