Properties of the swelling-induced chloride current (I_Cl,swell) in cardiac tissue are well known at the macroscopic level. The current is implicated in cell-volume regulation and in pathologies such as myocardial ischemia and dilated cardiopathies. Its main characteristics are a selectivity sequence of I^– > Br^– > Cl^– > F^– > gluconate, outward rectification, and activation by cell swelling. While a macroscopic chloride current with similar properties has been identified in human atrial cardiomyocytes, no study has been realized at the single channel level. Myocardium specimens were obtained from patients undergoing cardiac surgery according to the European Community Council Directive. Sinus rhythm was present in all cases. Cells were obtained by both enzymatic and mechanical dissociation. Using the cell-attached and the inside-out configurations of the patch-clamp technique, we characterized the properties of an outwardly rectifying chloride current (ORCC) at the unitary level in freshly isolated human atrial cardiomyocytes. In inside-out configuration, under symmetrical conditions (in mM: 140 NaCl, 4.8 KCl, 1.2 MgCl₂, 10 glucose and 10 HEPES, with 1 and 1.8 CaCl₂ in the pipette and the bath, respectively), the current-voltage relationship shows an outward rectification, 76.5 ± 14.7 pS and 8.1 ± 2 pS in the positive and negative voltage range, respectively (mean ± S.E.M., n = 5). The channel was chloride selective (P_Cl/P_Na = 18) with the permeability sequence I^– > Br^– > Cl^– > F^– > gluconate (P_I/P_Cl = 2.31, n = 4; P_Br/P_Cl = 1.2, n = 5; P_F/P_Cl = 0.33, n = 4; P_gluconate/P_Cl = 0.27, n = 4). In our conditions neither [ATP]_i nor [Ca²⁺]_i had effect on channel activity. The channel activity was decrease by the classical Cl^– channels blockers. Intracellular NPPB (5-nitro-2(-3-phenolpropylamino)benzoic acid) at 1 and 10 μM, SITS at 1 mM and DIDS at 100 μM reversibly reduced the channel activity to 63.1 ± 12.2% (p<0.05, n = 5) and 52.3 ± 13.1% (p<0.01, n = 5), 28.6 ± 14.5% (p<0.01, n = 5) and 60.3 ± 12.2% (p<0.05, n = 5) of control, respectively. The glibenclamide was also effective in blocking ORCC. EC₅₀ was estimated at 21.5 μM. ORCC detection increased after phorbol-12 myristate, 13-acetate (PMA) treatment, a PKC activator. ORCC activity was detected in 20% of patches (n = 35) in control condition and in 40% (n = 80) after 10 min treatment with 500 nM PMA (p<0.05). On cell-attached patches, bathing cells with hypotonic solution (causing cell swelling) increased channel detection (3.2%, n = 31 vs 76.5%, n = 17; p<0.01). Thus, the ORCC recorded at the single channel level possesses properties of the macroscopic I_Cl,swell reported in human atrial myocytes and as such should be considered a major component underlying this current. The human atrial ORCC shows similar properties with ORCC described by Duan et al. (1997) in rabbit atria.