Plasma Refill Rate and Intradialytic Hypotension during Hemodialysis in Hospitalized Patients

Background: Fluid removal during hemodialysis (HD) is often complicated by acute illness and comorbid conditions that render patients susceptible to hemodynamic instability even with modest fluid removal goals. We previously developed a straightforward method to quantify the rate of plasma refilling during HD, plasma refill rate (PRR), and found that PRR was associated with intradialytic hypotension (IDH) among patients receiving outpatient HD. We sought to investigate whether PRR is associated with IDH in hospitalized patients. Methods: We performed a prospective study of hospitalized patients receiving HD for either acute kidney injury (AKI) or end-stage kidney disease (ESKD) at a single center. Hematocrit monitoring (using CritLine-IV®) and ultrafiltration data were used to calculate PRR during HD, defined as the interval ratio of plasma refill volume to ultrafiltration volume, as previously established. Vital signs data, patient symptoms, and provider interventions were collected prospectively throughout HD. Cox proportional hazard regression was used to examine the relationship between PRR in the first 10 minutes of each HD session and time to IDH. Results: In interim analysis, we analyzed data from 99 patients with 1 to 3 HD sessions per patient. HD was performed for ESKD in 70% and AKI in 30% or participants. Mean pre-dialysis systolic blood pressure was 136.9±22.9 mmHg, serum albumin was 2.8±0.6 g/dL, and ultrafiltration rate was 8.1±3.5 ml/kg/h. We found that PRR in the first 10 minutes of HD was associated with IDH even after adjusting for age, sex, and pre-dialysis blood pressure. Specifically, low (PRR <0.5) compared to adequate PRR ≥0.5 was associated with an increased risk of IDH in patients with acute kidney injury (HR 2.93; 95% CI:1.62, 5.29) and midodrine-dependence (HR 2.38, 95% CI:1.25, 4.52). Conclusion: Low PRR early during the HD session was associated with an increased hazard of IDH, particularly in patients with AKI and midodrine-dependence. These results highlight the potential role of PRR for monitoring HD in patients with acute illness who may be susceptible to fluid shifts. Further studies are needed to understand the dynamics of PRR throughout the HD session and determine whether PRR is modifiable. Funding: NIDDK Support