The mechanisms of the diuretic effect of sodium-glucose cotransporter 2 (SGLT2) inhibitor and its predictors in heart failure (HF) patients with coexisting type 2 diabetes mellitus (T2DM) remain under investigation. A total of 40 hospitalized HF patients with T2DM (68 ± 13years old, male gender 63%) were prospectively enrolled and received ipragliflozin at a dose of 50mg once daily after breakfast for at least 4 consecutive days. They underwent first-morning blood and urine tests, and 24-h urine tests before and after short-term ipragliflozin therapy. Daily urine volume significantly increased from 1365 ± 511mL/day on day 0 to 1698 ± 595mL/day on day 3 (P < 0.001), which resulted in significant decreases in body weight and plasma brain natriuretic peptide level. Changes in 24-h urine volume were strongly and independently correlated with changes in 24-h urine sodium excretion (r = 0.80, P < 0.001), but was not significantly correlated with those in 24-h urine sugar excretion (r = 0.29, P = 0.07). Lower concentration of first-morning urine sodium and higher loop diuretic dosage before ipragliflozin therapy were associated with urine volume increment with ipragliflozin therapy, and former retained its independent predictor (Odds ratio 0.96, 95% CI 0.93-0.99, P = 0.02). First-morning urine sodium ≤ 53mEq/L and baseline loop diuretics ≥ 20mg/day predicted increased urine volume on day3 with high diagnostic accuracy. Ipragliflozin has acute natriuretic activity, and first-morning urine sodium and baseline dosage of loop diuretics strongly predicted the diuretic effects. Ipragliflozin therapy may restore responsiveness to loop diuretics in symptomatic HF patients with T2DM.