Right ventricular function across different non-invasive respiratory support strategies in severe community-acquired pneumonia.

BACKGROUND

Right ventricular (RV) systolic impairment is clinically relevant in acute respiratory failure, but its relationship with respiratory support modality in severe community-acquired pneumonia (CAP) outside the intensive care unit (ICU) is poorly defined.

OBJECTIVES

To compare RV longitudinal systolic performance and hospital length of stay (LOS) among patients receiving non-invasive ventilation (NIV), high-flow nasal cannula (HFNC), or Venturi-mask oxygen therapy.

METHODS

In this single-centre prospective observational study, adults with severe CAP (Pneumonia Severity Index class IV-V) requiring high-concentration oxygen or non-invasive respiratory support were enrolled. The modality was selected in the emergency department according to the treating physician's clinical and gas-exchange assessment. Echocardiography was performed shortly after ward admission while patients remained on the initial support modality.

RESULTS

Seventy patients were included (NIV, n = 21; HFNC, n = 22; Venturi mask, n = 27). Baseline demographic and cardiometabolic characteristics were comparable across groups. TAPSE and lateral tricuspid annular systolic velocity (S') showed a graded pattern, with the lowest values during NIV, intermediate values during HFNC, and the highest values with the Venturi mask (p < 0.001 for both). Mean LOS was 27.8 ± 5.0, 27.4 ± 2.9, and 15.2 ± 4.0 days, respectively (p < 0.001). In multivariable analysis, age and respiratory support modality were associated with LOS.

CONCLUSIONS

RV systolic performance and LOS differed across clinically selected respiratory support groups. These associations may reflect clinical severity rather than device-related effects. These findings reinforce the often-overlooked clinical importance of RV assessment in acute respiratory failure, extending observations from ARDS cohorts to severe CAP outside the ICU.