Page 69 - JSOM Winter 2024

FIGURE 1 Didactic phase lighting conditions for training in After models were on VV ECMO, participants managed
(A) set-up and priming of the veno-venous extracorporeal membrane ECMO-related complications and emergencies that were ini-
oxygenation circuit and (B) cannulation. tiated by a reviewer in the same fashion for all iterations and
performed damage control surgery and resuscitation proce-
dures. 27,28 Each team performed two iterations of testing with
white lights on and in darkness under similar conditions as di-
dactic training. Successful completion was determined by two
reviewers individually observing participants. A third reviewer
was present to adjudicate divergent scores. Each reviewer had
extensive ECMO cannulation and management experience in
civilian hospitals. Finally, participants completed a post-course
self-assessment.
(A) (B)
Animal Preparation
questionnaire based on questions from the Extracorporeal We sought to provide validation using a live animal model
Life Support Organization (ELSO) Specialist Training Man- given the inability to evaluate mortality and physiology using
ual focusing on cognitive (9 questions), technical (14 ques- a mannequin model. Practicing ECMO procedures on human
tions), and behavioral (2 questions) aspects of VV ECMO. volunteers is impractical and unethical; therefore, due to the
24
The pre- and post-test was the same, and participants did not similarities between human and swine cardiopulmonary anat-
know the content of the knowledge assessment prior to com- omy, we elected to use a total of 4 Yorkshire swine (Sus scrofa)
pleting it on arrival to the course nor that they would be com- weighing 40–70kg. These swine were fasted overnight except for
pleting the same assessment at the end of the didactic portion water ad lib. Swine were anesthetically induced with intramus-
of the course. A score of 80% was considered a passing score cular (IM) Telazol (4.4mg/kg) plus acepromazine (0.05mg/kg
based on the development of a previous ECMO knowledge IM). Following anesthetic induction, animal models were given
25
exam. For validation testing, several benchmarks were used atropine sulfate (0.05mg/g IM) and buprenorphine (0.02mg/kg
to measure success in performing tasks in a time-sensitive IM), and endotracheal intubation was performed. Isoflurane
manner. Previous studies have demonstrated that VV ECMO anesthesia was delivered using the A.D.S 2000 Veterinary An-
cannulation can take between 45 to 60 minutes from the start esthesia Delivery System a microprocessor controlled anes-
of the procedure to initiating VV ECMO. Time to circuit thetic ventilator. Following induction of anesthesia, vital signs
26
priming and set-up takes approximately 20 to 30 minutes, were continuously monitored using a Surgivet Advisor Tech
according to unpolished institutional data of novice trainees. monitor with 3-lead ECG, standard SPO , end-tidal CO , and
2
2
Demographic information was collected at the beginning of rectal temperature probe. Thermal support was provided us-
the course. Total instructional time at STC was 17 hours over ing a 3M Bair Hugger Warming Blanket System and isotonic
2 training days. intravenous (IV) fluid support (0.9% NaCl) was delivered via
an ear vein IV catheter.
Validation (Testing) Phase
Course participants then traveled to the 81st MDG CRL at To prevent cardiac dysrhythmias during guidewire placement,
Keesler Air Force Base (AFB). Participants were divided into animal models were given a 2–3mg/kg lidocaine bolus prior
two teams and tested on their ability to set up and prime the to cannulation and maintained on a 50µg/kg/min lidocaine
VV ECMO circuit and place Yorkshire swine (Sus scrofa) on constant-rate Infusion (CRI) until cannulation was complete.
VV ECMO using the provided checklists. A 2-hour timeframe Following confirmation of successful VV ECMO initiation
was allotted per testing event. Testing was marked complete and swine stabilization, animal models were given a 100mg
when all objectives were complete. bolus of ketamine and switched from gas anesthesia to total
intravenous anesthesia (TIVA) CRI (ketamine 12–48mg/kg/hr
Participants cannulated models for VV ECMO via the bi- and midazolam 0.5–1.5mg/kg/hr). Prior thoracotomy proce-
femoral approach. Two surgeons, one emergency physician, dures animal models were transitioned back to gas anesthesia.
and one anesthesiologist performed one access, dilation, and Upon completion of all VV ECMO training procedures animal
cannulation. Teams determined who would perform the ac- models were euthanized with an overdose (1mL/10 pounds)
cess and cannulations based on who would be most likely to of Euthasol (sodium pentobarbital 390mg/mL and phenytoin
perform these skills in the operational environment. Cannu- 50mg/mL) given intravenously or via intracardiac injection.
lation was accomplished using ultrasound-guided access and
percutaneous Seldinger technique. Participants had the option Data Storage and Analysis
of using a 5-French micropuncture kit (G48007, Cook Med- All participants were volunteers who signed consent to partic-
ical, Bloomington, IN) or using a needle in the dilator kit. A ipate in research prior to the course start. Study data were col-
guidewire was placed and serial dilations up to a 16-French lected and managed using Research Electronic Data Capture
dilator of the femoral veins were performed using a percuta- (REDCap) tools hosted at our institution. 29,30 Demographics,
neous insertion kit (LivaNova Sorin 8Fr-24Fr REF: 200-120). pre- and post-course self-assessments and pre- and post-course
The 19-French return cannula was placed in the right femoral knowledge assessments were collected from participants.
vein and the 19-French drainage cannula was placed in the left
femoral vein. The Maquet PVS 1938 cannula was used as the Data were analyzed with descriptive statistics. Parametric or
drainage cannula and the Medtronic Bio-Medicus 96670-019 nonparametric statistics were used based on the nature of the
was used as the return cannula. Cannulas were de-aired and data. Normality was assessed with the Shapiro-Wilk test
attached to the ECMO circuit through a wet-to-wet connec- and examination of stem-and-leaf as well as q-q plots. All
tion. The cannulas were then sutured in place for stabilization. data were normally distributed and presented as means and

V V ECMO Training Validation | 67