Patients and Methods
Patient Selection
This project was undertaken as a quality improvement initiative at
Baylor Scott & White Health, Temple, Texas. A retrospective analysis of
patient records was performed and the need for formal evaluation was
waived by our local IRB.
A retrospective review, from Jan 1st 2020 to July
31st 2021, of all isolated CABG patients from our
institution was conducted. All patients who underwent simplified UF
using Aquadex (Nuwellis, MN. USA) during their hospital course were
included in this analysis. Charts were reviewed using the electronic
medical records (Epic Systems, WI. USA) and patient’s demographics were
extracted and shown in Table 1 and 2. The exclusion criteria were ages
<18 years. The institution has adopted the cardiac enhanced
recovery after surgery (ERAS) protocols for all our cardiac surgery
patients. Hemodynamic monitoring (i.e., CVP, SvO2, CI, SVR) for goal
directed therapy was performed using the HemoSphere advanced monitoring
platform (Edwards Lifesciences, Irvine Ca).
The decision to perform simple UF in the post-operative period was made
by the attending cardiothoracic surgeons in collaboration with the CVICU
intensivists using our division’s protocol as summarized in Table 3.
Ultrafiltration Technique
Various modes of venous access were used including the previously placed
intra-operative ARROW MAC with ARROWg+ard Blue Technology 9F catheter
(Teleflex, USA, Fig 1), a peripheral dual lumen extended length catheter
(dELC) 6F/16Ga (Nuwellis, MN US, Fig 2) or a standard central venous
14Ga or 16Ga hemodialysis (HD) catheter (not shown). The approximate
time from UF request placed to initiation of fluid removal is typically
15 to 60 minutes once venous access is in place. Elective nephrology
consultations were submitted, and emergent consultations were reserved
for patients who met indications for hemodialysis. All diuretics were
discontinued once UF was in progress. Most patients were anticoagulated
with 400 to 600 Units Heparin per hour or Argatroban (to maintain PTT
40-50 secs) if heparin induced thrombocytopenia was suspected. The UF
rates of fluid removal varied from 100cc to 500/cc per hour at the
discretion of the surgeon, CVICU intensivist and ICU team using
clinical, hemodynamic and biomarkers to guide GDT. Intermittent CVP
readings were recorded from the central venous access line and goal
directed therapy was utilized with the HemoSphere advanced monitoring
platform (Edwards Lifesciences, Irvine Ca). The goal of UF was to target
a net negative 1-2L /day until CVP was <12, O2 requirements
were decreased, patient returned to baseline weight and or regained
adequate renal function to become diuretic responsive.
Statistical Methods
This was a retrospective study. Descriptive statistics were used
including sample size, mean, standard deviation, median, minimum, and
maximum for continuous variables and frequency and percent for
categorical variables.