Cardiac and hemodynamic effects of SGLT2 inhibitors in patients with cardiac amyloidosis

Background: Sodium glucose cotransporter type 2 inhibitors (SGLT2i) improve symptoms and prognosis in heart failure. However, their efficacy in patients with cardiac amyloidosis (CA) is unknown because these patients were excluded from SGLT2i trials. SGLT2i treatment altered hematocrit, hemoglobin, body weight and left ventricular filling pressures in previous studies. We studied the effects of SGLT2i treatment on these hemodynamic parameters in patients with cardiac amyloidosis.

Methods: The amyloidosis registry database of Leipzig university hospital (n=71) was screened retrospectively for patients with cardiac amyloidosis in whom an SGLT2i treatment was initiated and who had a clinical and echocardiographic follow-up (FU) within 12 months (n=13). Patients without SGLT2i treatment served as control (n=18). Primary endpoint variables were changes in hematocrit, hemoglobin, body weight and left ventricular (LV) end-diastolic pressure (LVEDP; measured from echocardiographic E/e’ ratio) from baseline (BL) to FU. Secondary endpoints were changes in LV ejection fraction (LVEF) and NT-proBNP.

Results: In the total cohort (n=31), mean age was 78±7years, 71% were male, 39% of patients were in NYHA functional class III and IV, median NT-proBNP was 2639pg/mL (25% - 75% interquartile range, 1289 – 8514). 48% and 16% of patients were in ATTR amyloidosis stage 2 and 3, respectively. Time between BL and FU was 6.3±2.5months. All but 4 patients were treated with tafamidis for cardiac amyloidosis. There was no difference in the baseline characteristics between groups, except for LV ejection fraction (51±13% SGLT2i vs. 60±11% control, p=0.041) and LVEDP (22±2mmHg SGLT2i vs. 21±3mmHg control, p=0.048). The dosage of diuretics did not change between BL and FU in both groups. Patients with SGLT2i showed increases in hematocrit (2.4±2.7% in SGLT2i vs. -1.5±2.3% in control, p=0.0002) and hemoglobin (0.6±0.8g/dl in SGLT2i vs. -0.6±1.2g/dl in control, p=0.0004) at FU compared to control (figure). SGLT2i treatment was associated with reductions in body weight (-2.5±3.1kg in SGLT2i vs. 0.3±2.8kg in control, p=0.023) and LVEDP (-1.9±1.3mmHg in SGLT2i vs. -0.1±3.5mmHg in control, p=0.025). LVEF increased (6.5±8.6% in SGLT2i vs. -2.8±9.1% in control, p=0.0054), while changes in NT-proBNP were similar between groups.

Conclusion: SGLT2i therapy was associated with improvements in cardiac and hemodynamic parameters in patients with cardiac amyloidosis. These results and their impact on clinical outcomes should be confirmed by larger prospective studies.