PURPOSE: Acute kidney injury (AKI) is common in the intensive care unit (ICU). Selected clinical studies have implied human atrial natriuretic peptide (hANP) improves renal function; however, the treatment effects for AKI are unclear. METHODS: A multicenter prospective observational study in 13 Japanese ICUs. The effects of hANP were estimated by the standardized mortality ratio weighted analyses of generalized linear models using propensity scores. The primary outcome was renal replacement therapy (RRT) or death in the ICU. RESULTS: Of 904 patients with AKI, 63 received hANP as a treatment for AKI. The primary outcome occurred in 20.5% (185/904). HANP did not reduce the risk of RRT or death in the ICU (risk ratio 1.12, 95% confidence interval [CI] 0.74 to 1.69) and was associated with a lower mean arterial pressure (β -3.8 mmHg, 95%CI -7.6 to -0.1), a longer hospital length of stay (β 12.0 days, 95%CI 1.2 to 22.8) and a lower eGFR at hospital discharge (β -10.4 mL/min/m2, 95%CI -19.1 to -1.7). No beneficial effect was observed in subgroups of cardiovascular surgery, sepsis, nor chronic kidney disease. CONCLUSIONS: In critically ill patients with AKI, the treatment effect of hANP was not evident on dialysis-free survival in the ICU.