The aim of the present study was to determine the use of static and dynamic haemodynamic parameters for predicting fluid responsiveness prior to volume expansion (VE) in intensive care unit (ICU) patients with systemic inflammatory response syndrome (SIRS).

We conducted a prospective, multicentre, observational study in 6 French ICUs in 2012. ICU physicians were audited concerning their use of static and dynamic haemodynamic parameters before each VE performed in patients with SIRS for 6 consecutive weeks.

The median volume of the 566 VEs administered to patients with SIRS was 1000mL [500–1000mL]. Although at least one static or dynamic haemodynamic parameter was measurable before 99% (95% CI, 99%–100%) of VEs, at least one them was used in only 38% (95% CI, 34%–42%) of cases: static parameters in 11% of cases (95% CI, 10%–12%) and dynamic parameters in 32% (95% CI, 30%–34%). Static parameters were never used when uninterpretable. For 15% of VEs (95% CI, 12%–18%), a dynamic parameter was measured in the presence of contraindications. Among dynamic parameters, respiratory variations in arterial pulse pressure (PPV) and passive leg raising (PLR) were measurable and interpretable before 17% and 90% of VEs, respectively.

Haemodynamic parameters are underused for predicting fluid responsiveness in current practice. In contrast to static parameters, dynamic parameters are often incorrectly used in the presence of contraindications. PLR is more frequently valid than PPV for predicting fluid responsiveness in ICU patients.