Dapagliflozin as an adjunct for ascites and glycemic control in patients with decompensated cirrhosis and diabetes: A retrospective cohort study

This is the authors' abstract. We don't add key points for this paper.

A BSTRACT Introduction: Evidence supporting sodium-glucose cotransporter 2 (SGLT2) inhibitors in decompensated cirrhosis remains limited. We evaluated dapagliflozin as an adjunct for ascites control and glycemic management in patients with decompensated cirrhosis and diabetes. Methods: This retrospective single-center cohort included 31 consecutive patients who started dapagliflozin 5–10 mg/day between January and October 2024 and were followed for six months. Ascites grade, large-volume paracentesis (LVP), diuretic dose, renal function, serum sodium, mean arterial pressure (MAP), liver biochemistry, and glycated hemoglobin (HbA1c) were assessed longitudinally. Results: Mean age was 48.5 ± 5.7 years and 27 patients were male. The distribution of mild: moderate: severe ascites improved from 0:14:17 at baseline to 22:9:0 at six months ( P < 0.001). LVP requirement decreased from 17/31 (54.8%) in the three months before treatment to 5/31 (16.1%) in the first three months after treatment ( P < 0.001), with no LVP during months 4–6. Furosemide and spironolactone doses decreased. Serum creatinine, sodium, and MAP remained stable; the paired six-month creatinine change was −0.09 mg/dL (95% confidence interval [CI], −0.23 to 0.04). HbA1c decreased from 7.68 ± 0.83% to 7.12 ± 0.58%; this was nominally significant but not after global Holm correction. No discontinuation due to renal dysfunction, hypotension, hyponatremia, or hypoglycemia occurred. Conclusion: In carefully selected patients with decompensated cirrhosis and diabetes, dapagliflozin was associated with improved ascites control without an observed renal, sodium, or hemodynamic safety signal. These findings are hypothesis-generating and do not support routine use until larger controlled studies establish efficacy and safety.