This was a prospective observational study of a convenience sample of adult ED patients (≥ 18 years old) planned to receive at least one resuscitation bolus of IV fluids by the treating team. The study was carried out at Leicester Royal Infirmary, an inner city acute hospital at Leicester, UK. The study was sponsored by the University of Leicester, UK and ethical approvals were obtained from Essex Research Ethics Committee (16/EE/0145).
Exclusion criteria were: clinical condition preventing the performance of a PLR (e.g. trauma), mental health presentations, alcohol intoxication, patients deemed unable to consent due to a pre-existing medical problem (e.g. dementia) and prisoners.
Patients were screened for eligibility by the clinical team or a member of the research team. Recruitment was carried out following a 2-stage consent process. Study procedure was initiated following initial verbal ascent (by patient, personal/professional consultee) following a brief explanation. This was followed by a full informed consent.
On recruitment, we recorded demographic data, presenting complaint, past medical history, and pre-inclusion fluid administered (including pre-hospital). Patients were followed up to record ED diagnosis, final discharge diagnosis, length of stay and mortality data. All biochemical results were recorded.
After the initial consent process standard monitoring was used to record standard physiologic parameters throughout the patients’ stay. From the standard usual care monitoring system the following parameters were recorded at baseline, and at the end of the study:
The TEB monitoring electrodes were applied following manufacturers recommendations - two to each side of the neck and two to each side of the patient’s lower thorax. The study procedure was carried out as follows:
-
Patients were monitored for at least 3 minutes in a 45° head-up position (baseline recording)
-
A 3-minute PLR test was carried out by trolley manipulation for up to 45° (PLR1)
-
Return to baseline position for at least 3 minutes (baseline 2)
-
Fluid bolus administration (typically 500mL over 15 minutes)
-
PLR was repeated following at least 10 minutes from fluid bolus end (PLR2)
TEB data were transferred to Microsoft Excel (Microsoft Corporation, United States) and data on CO, SV and HR were extracted. A minute by minute average was calculated for each parameter across the whole time series. To evaluate the overall trend, each parameter was averaged for each minute across the study cohort. To estimate preload responsiveness we considered the haemodynamic variables at seven time periods:
-
Baseline1: the minute immediately before leg raise
-
PLR1: the middle 1-minute of PLR1
-
Baseline2: the minute immediately before fluid bolus
-
FB: immediately following the end of fluid bolus
-
Baseline 3: 10 minutes following the end of fluid bolus
-
PLR2: the middle 1-minute of the second PLR2
-
Baseline4: at least 2 minutes following PLR2
Preload responsiveness was defined as ≥ 15% increase in SV. Accordingly, patients were classified into PLR+/PLR- (responders and non-responders to PLR), and R/NR (responders and non-responders to fluid bolus).
Descriptive data were presented as means with 95% confidence interval (CI), medians with interquartile ranges (IQR) and proportions as appropriate. Data analysis was performed using Graphpad Prism 7 (California, United States). Correlation between the stroke volume changes (∆SV) with PLR1 and FB was calculated using Pearson’s r. Categorical agreement between PLR1 and FB was evaluated using Cohen Kappa. Parametric and non-parametric tests were used as appropriate to evaluate statistical significance of baseline variables between responders and non-responders.
Patient and public involvement
Patients and public were involved in the design stage through a meeting with the Leicester Cardiovascular Research Review Group. The research protocol was presented to the group and the burden of the intervention was discussed. Feedback from the group was considered during the design and conduct of the study.