Abstract Background Few data exist on high flow nasal cannula (HFNC) use in patients with acute respiratory failure (ARF) admitted to general wards. Rationale and objectives To retrospectively evaluate feasibility and safety of HFNC in general wards under the intensivist-supervision and after specific training. Methods Patients with ARF (dyspnea, respiratory rate-RR > 25/min, 150 < PaO₂/FiO₂ < 300 mmHg during oxygen therapy) admitted to nine wards of an academic hospital were included. Gas-exchange, RR, and comfort were assessed before HFNC and after 2 and 24 h of application. Results 150 patients (81 male, age 74 [60–80] years, SOFA 4 [2–4]), 123 with de-novo ARF underwent HFNC with flow 60 L/min [50–60], FiO₂ 50% [36–50] and temperature 34 °C [31–37]. HFNC was applied a total of 1399 days, with a median duration of 7 [3–11] days. No major adverse events or deaths were reported. HFNC did not affect gas exchange but reduced RR (25–22/min at 2–24 h, p < 0.001), and improved Dyspnea Borg Scale (3–1, p < 0.001) and comfort (3–4, p < 0.001) after 24 h. HFNC failed in 20 patients (19.2%): 3 (2.9%) for intolerance, 14 (13.4%) escalated to NIV/CPAP in the ward, 3 (2.9%) transferred to ICU. Among these, one continued HFNC, while the other 2 were intubated and they both died. Predictors of HFNC failure were higher Charlson’s Comorbidity Index (OR 1.29 [1.07–1.55]; p = 0.004), higher APACHE II Score (OR 1.59 [1.09–4.17]; p = 0.003), and cardiac failure as cause of ARF (OR 5.26 [1.36–20.46]; p = 0.02). Conclusion In patients with mild-moderate ARF admitted to general wards, the use of HFNC after an initial training and daily supervision by intensivists was feasible and seemed safe. HFNC was effective in improving comfort, dyspnea, and respiratory rate without effects on gas exchanges. Trial registration This is a single-centre, noninterventional, retrospective analysis of clinical data.