Bedside Assessment of Right Ventricular Output Lying in Prone Or Supine

Recruitment and consent:

Patients will be screened for participation no later than the day before surgery. Potential participants will be given written and verbal information regarding the study and time to ask questions and consider their participation.

Randomization:

After end of surgery, at baseline in the ICU, patients will be randomly allocated (1:1 ratio) to supine position and prone position. Block randomization with variable block sizes will be performed with sequentially numbered, opaque, sealed envelopes. All performed by Statistiska konsultgruppen. Patient stratification according to cardiac index at end-surgery will be performed to ensure similar allocation of patients with cardiac index <2.2 L/min/m2 in each group. The investigator who enrolls study participants will not have access to the random allocation sequence.

Sample Size:

Based on previous studies of cardiac surgery or prone position an a priori sample size calculation was performed using a two-sample t-test comparison of means, with equally sized independent groups. With intention to detect at least a 20% difference in cardiac index in the intervention group and with 80% power, 0.05α, estimated mean CI 2.6(0.9 SD) in control group, and CI 3.2(0.9 SD) in the prone group, the estimated sample size was n=70, divided in 2 equally large groups. As patients intermittently require recurring surgery due to i.e. bleeding, patients might be excluded after inclusion. Enrolment continues until 70 participants have completed the entire study protocol.

Interim analysis:

Given uncertainty regarding expected baseline cardiac index and its standard deviation in the study population, an interim sample size reassessment will be performed after inclusion of 40% of the planned sample (n=28). The interim analysis will be limited to re-estimation of mean and standard deviation of baseline cardiac index, and will not assess treatment effects. If the observed assumptions differ substantially from those used in the original sample size calculation, the required sample size may be adjusted to preserve the planned statistical power of the study.

Description of method:

In included patients, anesthesia will be performed with Propofol, Fentanyl and Sevoflurane. As per clinical routine patients will have a central venous catheter, 1-2 arterial catheters, TEE probe and endotracheal tube inserted. Study participants will also get a pulmonary artery catheter (PAC) inserted prior to surgery to facilitate measurement of cardiac output using thermodilution and PVR. Body surface area is calculated using the Mosteller formula.

During surgery, after CPB is weaned, the attending anaesthesiologist is instructed to refrain from additional recruitment maneuvers and a PEEP >8 cmH2O. Unless FiO2 is >80% to maintain adequate oxygenation of the patient or the anaesthesiologist deems it necessary.

When surgery is finished, the patient will be moved to the Thoracic ICU. Ventilation performed with a Maquet Servo-U (Getinge, SE) is set to volume control, tidal volumes of 6 ml/kg IBW with a respiratory rate of 12-18 and PEEP 8 cmH2O. Data collection occurs in all patients in supine position (0). Hereafter the study group is turned to prone while the control group remains in supine position.

60 minutes after proning data is again collected (t1), whereafter a recruitment maneuver is performed by gradually increasing PEEP during 2,5 minutes in 6 steps to 20 cmH2O during. After 30 seconds PEEP is decreased in 6 steps during 2,5 minutes to reach PEEP 8 cmH2O. Data is collected after the recruitment maneuver (t2) and one hour after (t3), adding up to a total of 2 hours and 15 minutes in supine or prone position. The patients in prone position are turned to supine position and the last data collection occurs (t4). In total 5 time points for data collection. Postoperative care will now continue as per routine care. Patients will be extubated as soon as possible according to clinical routine and eventually discharged from the ICU.

Since a recruitment maneuver transiently increases right ventricular afterload there is a possibility not all patients will tolerate the recruitment maneuver as described above. As per clinical practice the level of airway pressures during recruitment will be adjusted when needed to avoid adverse effects.

Discontinuation of the intervention:

In order to minimize harm the following set criteria will be used to discontinue prone position:

• Hemodynamic instability with increased Norepinephrine >0.7mcg/kg/min
• Sustained arrythmia which requires treatment including DC conversion
• Any condition which, in the judgement of the investigator, might increase risk of harm to the patient

Blinding:

After allocation to supine or prone position neither the investigator, nor ordinary ICU personnel, will be blinded due to safety and impracticalities. Patients are blinded due to being under general anaesthesia during randomization until weaning from mechanical ventilation (after the study protocol is completed).

Data collection:

A site investigator will record data using a software developed by Getinge from the respirator to a designated research computer. All other data and variables, mentioned above, will be collected from the participants digital chart and real time monitoring (Philips IntelliVue \& BD Hemosphere).

Data management:

Collected data is pseudonymized and stored on a password protected designated research computer. Only authorized study personnel have access to the computer.

Original records, including signed written consent, are kept with strict confidence. These records will be kept in a secure, locked and limited access location at Sahlgrenska Univeristy Hospital during the study and after completion for an additional 10 years. At inclusion each patient is pseudonymized by receiving a participant ID, which is used in all data files.