Reference	Study design and population	Duration of study	Form of copper doses	Safety‐related parameters investigated	Summary of results
(Dassel de Vergara et al.,  1999 )	Cross‐sectional 956 households with infants and drinking water delivered in Cu pipes	NA	836 households (87.4%): water samples < 0.5 mg/L Cu 83 households: water samples ≥ 0.5 mg/L (including 38 households: ≥ 0.8 mg/L; max 2.6 mg/L)	Subsample of eight infants who were breast‐fed for up to 12 weeks or received ≥ 200 mL tap water/day with ≥ 0.8 mg Cu/L during their first 12 months: Examination of the liver by palpation and ultrasound Blood samples (serum Cu, caeruloplasmin, immunoglobulins (IgG, IgM, IgA), transaminases (GOT, GPT), GGT, total bilirubin and CRP)	Liver disease: liver palpation and ultrasound revealed no sign in any child Serum copper values above normal, CRP slightly above normal range: one infant at 8 months; other parameters in the norm
(Dieter et al.,  1999 )	Retrospective study 103 cases of early childhood cirrhosis in Germany	Between 1984 and 1994	Cu content of drinking water	Histologically confirmed early childhood cirrhosis	5 cases considered as probably related to chronic and excessive intake of copper (coincided with high hepatic copper contents and copper plumbing/acid well water); 9–26.4 mg Cu/L in water
(Scheinberg and Sternlieb,  1994 )	Case reports 7 children (< 2 years) with non‐Indian childhood cirrhosis	NA	Cu content of drinking water: 0.05 to 6.8 mg Cu/L		Cirrhosis; evidence of a genetic aetiology in three of the seven infants
	Retrospective study Children (0–5 years) from three Massachusetts towns	64,124 child‐years of exposure (between 1969 and 1991)	Cu content of drinking water: 8.5–8.8 mg Cu/L	Records from Massachusetts Department of Public Health	No deaths from cirrhosis or any form of liver disease
(Zietz et al.,  2003 )	Cross‐sectional 2,944 households with infants (Berlin area)	NA	Cu content of drinking water: Composite sample type 1 (aliquots of 100 ml of tap water collected each time it was used in the household): mean 0.44 mg/L; max 3.5 mg/L Composite sample type 2: collection of 250 ml of tap water in the morning, at noon, in the evening and before going to bed: mean 0.56 mg/L; max. 4.2 mg/L 0.8–4.2 mg Cu/L	Infants from families having a Cu concentration ≥ 0.8 mg/L in water samples (29.9% of all sampled households) and who had ingested ≥ 200 mL tap water/day for at least 6 weeks were recommended to undergo a paediatric examination (541 infants eligible): 517 infants were inspected and examined by a physician 183 received a paediatric examination (liver palpation and ultrasound imaging) and blood serum analysis (serum Cu; caeruloplasmin; IgG, IgM, IgA; GOT; GPT; total bilirubin; CRP)	No sign of liver dysfunction (serum GOT, GPT, GGT and serum copper outside the reference range in eight cases of which six had clinically diagnosed infection and one had a liver haemangioma and a ureteric obstruction; abdominal ultrasound imaging slightly unusual in five cases considered likely caused by infection) No signs of a negative health effect found in dose–response analyses of daily and total copper intakes of the infants from tap water and serum GOT, GPT, GGT, total bilirubin, serum copper or caeruloplasmin
(O'Donohue et al.,  1999 )	Case report 1 subject (26 years)	42 months	30 mg Cu/day for 30 months followed by 60 mg Cu/day for 1 year	Blood and urine samples Physical symptoms	Admitted to hospital after 6‐week history of malaise, jaundice and abdominal swelling Acute renal failure Severe liver cirrhosis, necessitating liver transplantation The explanted liver had a copper conc. of 3.230 mg/g dry weight; zinc conc. was normal

CRP: C‐reactive protein; GGT: γ‐glutamyl transferase; GOT: glutamic‐oxaloacetic transaminase; GPT: glutamic‐pyruvic transaminase; IgA: immunoglobulin A; IgG: immunoglobulin G; IgM: immunoglobulin M; NA: not available.