This prospective cohort study aim to investigate the ability of multiple types of assessments including 1) the modified Venous Excess Ultrasound (VExUS) assessment, 2) non-invasive estimation of absolute blood volume (ABV), and 3) change in carotid Doppler at the start of IKRT to predict IDHE in acutely ill hospitalized patients. The secondary aim will be to determine whether each modality improves the prediction of IHDE compared to the evaluation of the attending physician and whether they also predict cerebral hypoxia during IKRT measured by near-infrared spectroscopy (NIRS). Finally, detailed hemodynamic data including relative blood volume monitoring, tissue oximetry, and other parameters will be collected continuously during IKRT sessions enabling exploratory analyses aimed at identifying hemodynamic phenotypes related to IDHE during IKRT.

Background:

Fluid accumulation and intra-dialytic hypotensive events (IDHE) are independently associated with worse prognosis in patients undergoing intermittent kidney replacement therapy (IKRT), both in the setting of acute kidney injury (AKI) and in patients with end-stage kidney disease (ESKD). An optimal IKRT prescription needs to achieve adequate fluid homeostasis while mitigating the risk of IDHE. However, predicting tolerance to fluid removal is challenging, particularly in acutely ill hospitalized patients in whom intravascular volume and compensatory mechanisms are affected by the presence of acute illness.

The following modalities will be investigated in this study to determine if they predict IDHE during hemodialysis in hospitalized patients:

• The modified VExUS assessment before the hemodialysis session
• Estimation of absolute blood volume during the hemodialysis session
• Removed in protocol version 2.0: Change in carotid Doppler parameters in response to fluid administration at the start of hemodialysis

Objectives:

Primary objective:

To determine whether the studied modalities accurately predicts intradialytic hypotension events (IDHE) in hospitalized patients, including critically ill and non-critically ill patients

Secondary objectives:

To determine whether the studied modalities predict cerebral desaturations events measured by near infra-red tissue oximetry (NIRS) during IKRT To determine if adding information from the studied modalities result in significant improvement in the risk prediction of IDHE when added to the subjective assessment of the attending clinician

Study design:

A cohort of hospitalized patients undergoing hemodiafiltration with ultrafiltration will be monitored during two separate dialysis sessions. The following markers will assessed:

1. Absolute blood volume, estimated based on the change in relative blood volume measured by the Fresenius BVM after infusion of a replacement fluid bolus of 240 mL given 15 minutes after dialysis initiation using the Kron et al method.
2. The modified VExUS from the Doppler assessment of the portal, hepatic, femoral, and splenic vein before the hemodialysis treatment.

Inclusion Criteria:

1. Adult patient ≥18 year-old
2. Hospitalized at the CHUM in the intensive care unit (ICU) or general ward
3. Receiving or planned ≥3x/week IKRT treatments
4. Planned on-line hemodiafiltration (HDF) or hemodialysis (HD) sessions with ≥1L of fluid removal per session

Exclusion Criteria:

1. Planned hospital discharge, death, or transition to comfort care within 48 hours according to the attending physician
2. End of active care (awaiting hospital discharge)

Hospitalized adults with acute kidney injury or chronic kidney disease requiring hemodialysis with on-line hemodiafiltration.

Will require data sharing agreement