The response of serum alkaline phosphatase (AP), a zinc-dependent metalloenzyme, to zinc administration via the dialysate (400 µg/l) was examined in 14 hypozincemic ( < 30th percentile of dialysis patients) hemodialysis patients and in 14 placebo-treated matched dialysis control patients. Plasma zinc and serum AP were measured three times: prior to, once weekly during (5 weeks), and 2 weeks after addition of zinc to the dialysate. The serum zinc levels remained stable in placebo-treated controls (initial 87.7 ± 12.5; final 78.6 ± 8.3 µg/dl) and increased in zinc-treated patients (initial 76.4 ± 8.3; 5th week 96.9 ± 13.3; 2 weeks after zinc withdrawal 82.3 ± 12.2 µg/dl). There was a slight increase of AP with time in placebo controls (initial 90.2 ± 26.5; 5th week 100 ± 29 U/l) and a more pronounced increase in zinc-treated patients (initial 90.8 ± 19.9; 5th week 113 ± 20.9 U/l). The difference between the two groups was marginally significant (p < 0.05; analysis of variance). It is concluded that zinc repletion via dialysate with documented increase of serum zinc levels in initially hypozincemic dialysis patients causes a reversible increase of serum AP. The result is compatible with some tissue zinc deficiency in hypozincemic dialysis patients.