Study Hypothesis
The aim of this study is to demonstrate that circulatory arrest using mild hypothermia (32±1°C) and uSACP will result in a 15% absolute risk reduction (from 35% to 20% - see sample size section for calculation and Table 1) in our composite outcome during aortic hemiarch surgery, when compared to moderate hypothermia (26±1°C).
Study Design
This trial is a prospective multi-centre, single-blind two-arm RCT comparing mild versus moderate hypothermia for circulatory arrest in hemiarch surgery on a composite outcome of neurological and acute kidney injury in 282 patients undergoing aortic hemiarch surgery with uSACP. Consenting adult patients (≥ 18 years of age) undergoing ascending aorta and hemi-arch replacement with uSACP and an anticipated circulatory arrest time of less than 20 minutes will be randomized 1:1 to moderate hypothermia (26±1°C, Control Group) versus mild hypothermia (32±1°C, Treatment Group).
Study Setting
The principal study site for this RCT will be the University of Ottawa Heart Institute (UOHI), which is a quaternary care cardiovascular centre with a large thoracic aortic program, performing over 200 thoracic aortic operations per year. Other participating sites in Canada include: the Foothills Medical Centre (University of Calgary), London Health Sciences Centre (Western University), and Quebec Heart and Lung Institute (University of Laval).
Study Description
Eligibility Criteria
Inclusion Criteria
- Age ≥ 18 years
- Planned unilateral or bilateral selective anterograde cerebral perfusion
- Anticipated lower body arrest time of < 20 minutes
- Able to provide written informed consent
Exclusion Criteria
- Surgery for aortic dissection or urgent/emergent operations
- Total arch replacement
- Inability to perform unilateral selective anterograde cerebral perfusion (severe axillary or innominate artery atherosclerosis/stenosis)
- Patients with known/documented coagulopathies
- Patients with cold agglutinin disease or those that test positive on routine preoperative screening
- Pre-existing severe neurological impairment or inability to accurately assess neurocognitive function as determined by the operating surgeon
- Severe carotid disease
a. Any patient with previously documented carotid stenosis of ≥ 70% (via Doppler ultrasound, magnetic resonance angiography, or computer tomography angiography) without neurological deficits
b. Carotid stenosis ≥ 50% with neurological deficits
c. Previous carotid endarterectomy or stenting
- Patients in renal failure or currently being treated with renal replacement therapy (RRT) or estimated glomerular filtration rate (eGFR) < 30 ml/min/1.73m2
- Use of an investigational drug or device at time of enrollment
- Participation in another clinical trial which interferes with performance of this study’s procedures or assessment of outcomes
Enrollment
Screening will occur at the time of the initial appointment, at which time patients will be seen and assessed by the cardiac surgeon for consideration of aortic arch surgery. Patients who are booked for aortic hemiarch surgery with planned hypothermic circulatory arrest and ACP will be assessed for the aforementioned eligibility in the clinical trial. Once deemed eligible and written informed consent obtained, baseline neurological screening will be performed by a trained personnel.
The following baseline information will then be collected at either this initial visit, or at the time of the preoperative anesthesia clinic visit (see SPIRIT Figure 1):
- Clinical Information
a. Demographics – age, sex
b. HTN (Hypertension)
c. DM (Diabetes)
d. DLP (Dyslipidemia)
e. NYHA Class (New York Heart Association)
f. Aortic Stenosis
i. Graded: Mild, Moderate, Severe
g. Aortic Valve Regurgitation
i. Graded: 1 to 4+
h. CCS Class (Canadian Cardiovascular Society)
i. CAD (Coronary Artery Disease)
j. Previous history of stroke
k. Previous history of TIA (Transient Ischemic Attack)
l. Carotid Stenosis
m. Renal Disease
n. PVD (Peripheral Vascular Disease)
o. COPD (Chronic Obstructive Pulmonary Disease)
p. Pulmonary HTN
q. Smoking history
r. ETOH use (Ethanol)
s. Congenital aortic disease
t. Aortic Diameter
u. Euroscore II
- Physical Examination
a. Height (cm)
b. Weight (kg)
c. BMI (kg/m²) – Body Mass Index
d. BSA (m²) – Body Surface Area
- Laboratory Tests/Investigations
a. Hb (g/L) - Hemoglobin
b. Plt (x109/L) – Platelet
c. INR – International Normalized Ratio
d. Cr (umol/L) - Creatinine
e. BUN (mmol/L) – Blood Urea Nitrogen
f. eGFR (ml/min/1.73m2) – estimated glomerular filtration rate
g. HbA1C (%) – Hemoglobin A1c
- Neurocognitive Testing
a. NIHSS (The National Institutes of Health Stroke Scale)
b. MOCA (Montreal Cognitive Assessment)
c. MRS (Modified Rankin Scale) – only postoperatively in event of stroke
d. BI (Barthel Index)
- Quality of life
a. SF-12 (Short Form) survey
Randomization
Once study eligibility, consent, and baseline assessment are completed, subjects will be randomized using a web-based randomization tool (DACIMA 3.3.8) (21) in a 1:1 ratio on the day of surgery to either mild or moderate hypothermia strategies during aortic hemiarch arch surgery with uSACP.
In order to ensure a normal distribution of kidney function among patients in both treatment arms, stratified randomization will be performed based on two eGFR ranges: 30-59 ml/min/1.73m2 and ≥ 60 ml/min/1.73m2. (22) Randomization will also be stratified by site to reduce possible confounding from differences in standard of care practices across sites.
Block randomization to each treatment arm will be used across participating centres.
Study Intervention
Subjects will be randomly allocated to one of the following two arms:
- Mild hypothermia (32°C±1°C) strategy for circulatory arrest (intervention group)
- Moderate hypothermia (26°C±1°C) strategy for circulatory arrest (control group)
Conduct of Aortic Hemiarch Surgery and Hypothermic Strategy
Sequence of operation up until circulatory arrest with CPB will be the at surgeon’s discretion. All surgical procedures will be performed via median sternotomy (or upper hemi-sternotomy). During the aortic arch anastomosis, continuous, unilateral selective ACP using axillary or innominate artery cannulation will be employed. Unilateral SACP may be converted to bilateral ACP at the surgeon’s discretion if adequate cerebral flows are not achieved, or if there are concerns with cerebral oximetry measurements.
Once on CPB, the patient will be cooled to a nasopharyngeal (NP) temperature of either 32°C or 26°C, depending on to which treatment arm the patient has been randomized. Bladder temperature and venous blood temperature will both be monitored as additional temperature sites. Unilateral SACP will only be initiated once the target temperature has been reached. ACP via the axillary or innominate artery will be commenced with target flows of 10 – 15 ml/kg/min with temperatures of 32°C or 26°C. ACP flows will be titrated to a right radial blood pressure (BP) of 60 – 70 mmHg. Heater/cooler temperatures are not to exceed 37.5°C and a temperature gradient of less than 10°C should be maintained at all times.
After completion of the aortic hemiarch replacement, CPB will be resumed and the patient re-warmed to 35°C prior to coming off CPB, with a ≤ 1°C temperature difference between temperature monitoring sites(NP and bladder probes).
Transfusion Strategy
Transfusion triggers (for packed red blood cells only) will be adhered to whenever possible without compromising patient safety or centre specific transfusion protocols. Thus, a liberal transfusion threshold of Hb <95 g/L will be used intraoperatively and in the intensive care unit (ICU), while a Hb < 85 g/L will be adhered to after transfer to the ward. These values were chosen based on the TRICS-III trial (Transfusion requirements in Cardiac Surgery III) and will be used for the duration of the patients hospital stay. (23)
Frequency and Duration of Follow-up
Study participants will be followed up daily during their postoperative course in hospital, including the ICU. Intraoperative information will be collected from the anesthetic record, surgical notes, and perfusion records. Intraoperative data collection will include total operative time (skin to skin time), total CPB time, total cross-clamp time, total hypothermic cardiac arrest time, uSACP time, cooling time, re-warming time, nadir nasopharyngeal and bladder temperature, mean arterial systolic and diastolic BP, nadir hemoglobin concentration (g/L), nadir hematocrit (%), intraoperative red blood cell (RBC) transfusion (units), and highest dose/agent used for intraoperative inotrope or vasopressor support.
Postoperative data will include tabulations from the following areas:
- Mortality
a. Death from any cause
b. Up to post op day 60±14
- Neurological Injury
a. TIA
b. Stroke
c. NIHSS – POD 2±2, 6±1 (or prior to discharge), and 60±14
d. MOCA – POD 2±2, 6±1 (or prior to discharge), and 60±14
e. MRS– only in event of postoperative stroke
f. BI– only in event of postoperative stroke
- Delirium
a. Assessed and scored as per the Confusion Assessment Model (CAM) - POD 2±2 and 6±1
- Acute Kidney Injury
a. Creatinine and BUN – POD 0, 1, 2±2, 4±1, 6±1 (or prior to discharge), and 60±14
b. Urine output – up to 48 hrs
c. Renal replacement therapy (dialysis)
- Prolonged Mechanical Ventilation
a. Mechanical ventilation times ≥ 48 hrs
b. Measured in hours from time of admission to the ICU
- Coagulopathy
a. Mediastinal re-exploration for bleeding or tamponade
b. Chest tube outputs
- Perioperative Transfusions
a. Packed Red Blood Cells (pRBCs)
b. Platelets
c. Fresh Frozen Platelets (FFP)
d. Cryoprecipitate
- Postoperative Myocardial Infarction
a. Electrocardiogram (ECG)
b. Troponins
- Inotropic Support greater than 48 hours
a. Agent (amiodarone, epinephrine, norepinephrine, dopamine, dobutamine, milrinone, vasopressin) and dose (mcg/kg/min)
- Length of Stay
a. Intensive care unit (ICU) days
b. Total hospital days
- Quality of Life Measures
a. SF-12 – POD 60±14
Postoperative Assessment and Follow-up Study Visits
On postoperative days 2±2 days, 6±1 days, and 60±14 patients, will undergo neurocognitive screening by trained personnel. NIHSS and MOCA examinations will be made at each time point (or prior to discharge for time point 6±1 days). In the event a neurological deficit is identified during assessment and/or based on patient symptoms (focal/global, motor/sensory loss or prolonged delirium/agitation), neurological imaging will be obtained using Computed Tomography (CT) head or diffusion-weight magnetic resonance imaging (DW-MRI). MRIs will be used only if symptom onset is acute in nature requiring rapid diagnosis. Patients will also undergo Modified Rankin Scale assessment only in the event of a postoperative stroke. Barthel Index scoring will be performed preoperatively and postoperatively only in the event of a stroke. Quality of life measures will be assessed using SF-12 health survey and obtained preoperatively and at final follow-up at 60±14 days. (See SPIRIT Figure 1)
Primary Outcome and Definitions
The primary efficacy endpoint for this study will be a composite of neurologic and acute kidney injury that occurs at any time over the 60±14 day study period. Definitions are defined in secondary outcomes below.
Secondary Outcomes and Definitions
Secondary endpoints will include:
- Neurologic injury is divided into two major categories: Temporary Neurologic Dysfunction (TND) or TIA and Permanent Neurologic Dysfunction (PND) or stroke.
a. TND orTIA – neurologic symptoms lasting < 24 hours and without evidence of infarction
i. Neurological imaging has to be normal with resolution of all symptoms within 24 hours
b. PND or stroke – presence of either new focal (embolic stroke) or global (diffuse coma) deficits which persists for greater than 72 hours
i. Positive radiographic evidence of infarction in the appropriate territory
In the event of a suspected neurological injury the clinical team will be alerted and the study neurologists will be consulted. Neurologic injury will be quantified using a combination of validated cognitive screening tests and neurologic imaging. Screening tests will be performed preoperatively and on postoperative days 2±2, 6±1 (or prior to discharge), and 60±14 by trained personnel (study coordinator, study neurologist, study nurse practitioner). Note both BI and MRS scores will only be performed postoperatively in the event of a postoperative stroke and will occur at same times as the follow up NIHSS/MOCA testing. These points in time have been chosen based on previous studies (24) that actively monitor for stroke post cardiac surgery, as well as to allow for long-term follow up of patients post stroke. Screening tests include:
c. The National Institutes of Health Stroke Scale (NIHSS) – highly predictive of hospital disposition and long-term stroke outcomes. Scores range from 0 to 42. It has been shown for each 1-point increase in NIHSS, the likelihood of going home is significantly reduced. (25-27)
i. 0 = no stroke
ii. 1-4 = minor stroke
iii. 5-15 = moderate stroke
iv. 15-20 = moderate/severe stroke
v. 21-42 = severe stroke
vi. Study neurologists following serial NIHSS scores will determine whether there was a change in examination from previous exams and whether this change was because of a suspected stroke.
- For the purpose of this study, severe strokes are defined as NIHSS ≥ 10 (24)
- Clinically significant strokes are defined as a change in NIHSS ≥ 4
a. This is based on data which shows NIHSS scores less than 6 indicate a strong likelihood of hospital discharge, with good recovery, and no long term deficits (26,28)
d. The Montreal Cognitive Assessment (MOCA) – highly sensitive in detecting executive dysfunction. Scores range from 0 to 30. (29)
i. Scores > 26 – no cognitive impairment (normal exam)
ii. Scores < 26 – mild cognitive impairment
e. Modified Rankin Scale (MRS) – only to be used in the event of a postoperative stroke. The MRS is a reliable and reproducible scoring method for the assessment of deficits post stoke in patients. Scores range from 0 to 6. (30,31)
i. 0 - No symptoms.
ii. 1 - No significant disability. Able to carry out all usual activities, despite some symptoms.
iii. 2 - Slight disability. Able to look after own affairs without assistance, but unable to carry out all previous activities.
iv. 3 - Moderate disability. Requires some help, but able to walk unassisted.
v. 4 - Moderately severe disability. Unable to attend to own bodily needs without assistance, and unable to walk unassisted.
vi. 5 - Severe disability. Requires constant nursing care and attention, bedridden, incontinent.
vii. 6 - Dead.
f. Barthel Index of Activities of Daily Living (BI) – the BI is formed by several disability indexes and is a reliable and reproducible scoring method for the assessment of activities of daily living (ADLs) post stoke in patients. Scores range from 0 to 20.(32-34)
i. Lower scores indicate increased disability.
ii. When used to measure improvement after rehabilitation, changes of more than two points in the total score reflect probable genuine change.
iii. Change on one item from fully dependent to independent are also likely to be genuine.
2. Incidence of Acute Kidney Injury (AKI)
a. Postoperative AKI will be assessed using the KDIGO criteria (35-37) and summarized in table 1 below.
b. Creatinine measurements will serve as the main indicator for assessing AKI.
c. All those with KDIGO stage 1 or higher AKI will be considered to have sustained postoperative AKI.
d. Creatinine measurements will be performed with routine bloodwork on POD 0, 1, 2±2, 4±1, 6±1, and 60±14 days.
e. Urine output for the diagnosis of AKI will only be measured in the immediate postoperative period (48 hours) .
f. Blood urea nitrogen (BUN) levels will also be obtained on the same days.
Table 1: KDIGO Classification of AKI
Stage
|
Serum Creatinine
|
Urine Output
|
1
|
1.5-1.9 times baseline
or
≥ 0.3 mg/dl (26.5 µmol/l) increase
|
< 0.5 ml/kg/hr for 6-12 hours
|
2
|
2.0-2.9 times baseline
|
< 0.5 ml/kg/hr for ≥ 12 hours
|
3
|
3.0 times baseline
or
Increase in serum creatinine to ≥ 4 mg/dl (353.6 µmol/l)
or
initiation of renal replacement therapy
|
< 0.3 ml/kg/hr for ≥ 24 hours
or
Anuria for ≥ 12 hours
|
3. Incidence of Delirium
a. Delirium –reversible postoperative delirium lasting more than 48 hours without localizing signs
i. Delirium assessment will be performed using CAM
1. For a diagnosis of delirium by CAM, the patient must display:
a. Presence of acute onset and fluctuating course AND
b. Inattention
2. And either one of:
a. Disorganized thinking OR
b. Altered level of consciousness
ii. Assessments will be performed on POD 2±2 and6±1
4. Death
a. All cause postoperative 90-day or in-hospital mortality
5. Cardiopulmonary bypass time (CPB) [minutes]
6. Bleeding rates to qualify for mediastinal re-exploration
a. Indication for exploration defined as postoperative mediastinal bleeding of (38,39):
i. > 500 mls in any one hour
ii. > 1000 ml in any 4 hour period
iii. or at surgeons discretion
7. Incidence and quantity [# units/patient] of perioperative blood transfusions
a. Defined as all intraoperative and postoperative transfusions up to POD 7 or discharge (whichever comes first)
b. The transfusion of packed red blood cells (pRBCs) will adhere to the previously defined transfusion trigger strategy whenever possible without compromising patient safety
i. A liberal transfusion threshold of Hb <95 g/L will be used intraoperatively and in the ICU, while a Hb < 85 g/L will be adhered to after transfer to the ward.
c. Number of platelets, fresh frozen plasma (FFP), and cryoprecipitate will also be tabulated for the same time period
i. No transfusion triggers given and left to discretion of caring physician
8. Prolonged Mechanical Ventilation
a. Defined as those requiring ≥ 48 hours of mechanical ventilation
b. Measured in hrs from time of admission to the ICU
9. Perioperative Myocardial Infarction
a. Clinically diagnosed using a combination of electrocardiographic (new Q wave on 12 lead ECG) and biochemical (Troponin I > 45 ng/L) markers or both.
10. Length of Stay
a. Intensive care unit (ICU) hours
b. Hospital days
11. Quality of life (SF-12) survey
a. A measure of perceived health (health-related quality of life [QoL]) that describes the degree of general physical health status and mental health distress (40)
b. Will be assessed preoperatively and on POD 60±14
Statistical Consideration
In a prospective contemporary study of patients over the age of 65 undergoing aortic valve replacement, Floyd et al. observed an incidence of clinical stroke of 17% using similar assessment modalities to those proposed in this trial. (24) Interestingly, clinically documented strokes in the Society of Thoracic Surgeons (STS) adult cardiac surgery database in the same patients were only 7%, suggesting that careful and systematic documentation can reveal a ~ 2.5 fold higher incidence of neurologic injury than routine clinical evaluation. Furthermore, this study found increasing CPB to be an important risk factor for neurologic injury. (24)
In patients undergoing aortic arch surgery, the incidence of neurologic injury is expected to be higher due to a variety of reasons including: longer CPB times, manipulation of the aortic arch and branch vessels for cannulation or clamping, and injury associated with hypothermia and re-warming. In a large retrospective study of over 45,000 patients undergoing arch surgery, Hughes et al. found a clinical stroke rate of ~ 6.62%. Notably, the authors did not include patients who may have suffered TIAs or other temporary neurologic dysfunction. Their data was also obtained from the STS database, with no prospective or systematic neurologic evaluation. Thus for the purpose of this RCT, we hypothesize that patients in the moderate hypothermia group will experience a 15 - 20% incidence of neurologic injury.
With respect to the incidence of AKI, previously reported rates of stage 1 AKI in patients undergoing cardiac surgery by Boodhwani et al. have revealed an incidence of around 22.35% based on the Acute Kidney Injury Network (AKIN) criteria for staging AKI. (41) Their study population was a mixed surgical population with infrequent procedures such as heart transplantation, ventricular assist device placement, and complex congenital abnormality repair, being excluded.
This RCT will use the newer KDIGO criteria for assessing AKI, which identifies significantly more AKI in patients than both the Risk, Injury, Failure, Loss of kidney function, and End-stage kidney disease (RIFLE) and AKIN criteria. (36,42) Specifically, KDIGO criteria has been shown to have a greater sensitivity in diagnosing AKI when compared to RIFLE and AKIN (51% versus 46.9% for RIFLE, p <0.01 and 38.4% for AKIN, p<0.001). (42) It is also expected that the incidence of AKI will be higher in this study based on the type of surgery being performed, which necessitates a period of lower body ischemia time. Thus we hypothesize a 25% incidence of KDIGO stage 1 AKI for this trial.
Sample Size
Taking into consideration the aforementioned incidence rates of both neurologic and acute kidney injury, a composite outcome of ~ 35% should represent the incidence of these injuries in patients undergoing hemiarch surgery using traditional methods of hypothermia. This accounts for a 5 - 10% overlap that will likely exist, where a patient will suffer both types of injuries.
We hypothesize that circulatory arrest using mild hypothermia (32°C) and uSACP during aortic hemiarch surgery will result in a 15% absolute risk reduction in composite outcomes (neurologic and acute kidney injury) from 35% to 20%. With an alpha (type 1 error) of 0.05 and power of 80%, and a 5% loss to follow-up, approximately 141 subjects will be needed in each group for a total of a 282 patients.
Clinical Significance
Aforementioned, strokes post cardiac surgery have been shown to double the duration and costs of hospitalization, and are associated with increased mortality and severe physical disability. (12) Similarly, kidney injury resulting in end-stage renal disease is a major health problem due to its high morbidity and mortality, as well as social and financial implications. (43) These post-operative complications result in significant lifestyle morbidity, as well as added expenditures to the Health Care System.
With an ARR of 15%, the relative risk (RR) and RR reduction (RRR) would be 57% and 43% respectively, suggesting the proposed intervention carries benefit, and has a medium-sized “relative” effect, with only 7 patients needing to be treated with the mild hypothermic strategy to avoid an adverse event. Considering the proposed intervention adds no extra costs, with temperature being the only variable manipulated in the normal standard of care for these patients, the proposed intervention potentially carries both a significant and meaningful benefit to patients, while saving Health Care expenditures.
Statistical Analysis
The trial will be analyzed on a true Intention to Treat (ITT) basis, including all trial participants who were randomized regardless of adherence to treatment protocol, including those who are lost to follow-up or may have died. Patient crossover will be tracked and reported with the final trial results. In addition, a ‘modified ITT’ analysis will be performed (excluding patients not assessed for the primary outcome secondary to lost to follow-up, early death, non-adherence or other justifiable deviations) to account for potential biases introduced with ITT analyses.
The primary endpoint of our composite outcome will be evaluated using Chi Squared test (or Fisher’s Exact test if cell count is < 5 in any cell). Categorical secondary endpoints will be analyzed in the same method, while continuous secondary endpoints will be analyzed using Students T-test (or Wilcoxon Rank Sum Test, if the data are not normally distributed). Continuous variables will be reported as mean ± standard deviation or median (Interquartile Range [IQR]) for non-normally-distributed variables. Categorical variables will be reported as counts and percentages. As the primary outcome is a composite of both neurologic and acute kidney injury, secondary outcome analysis will pay particular attention to individual assessment of these indices for treatment effect and positive correlation.
Exploratory multivariable logistic regression analysis will be performed to determine risk factors for neurologic and renal complications post-operatively. A p-value of <0.05 will be considered statistically significant. Multivariable analysis will include variables with p<0.20 in univariate analysis in addition to clinically significant variables.
Once data collection is a complete for all study participants, a Statistical Analysis Plan (SAP) will be created prior to data analysis and release of results. As the principle statistician involved in this trial, Dr. George Wells (PhD) will primarily be responsible for this analysis. Dr. Wells, is the Director of the Cardiovascular Research Methods Centre at the UOHI and Professor in the School of Epidemiology, Public Health and Preventive Medicine at the University of Ottawa.
Data Collection – Case Report Forms (CRFs)
Data collection will be completed by authorized study personnel designated by the site investigator. Appropriate training will be completed with the site investigator and all authorized personnel prior to the study being initiated. Data collection started on paper, however, an electronic data collection system has been developed and is now used as the primary data collection method for all sites.
Monitoring and Auditing
Monitoring of study compliance and data collection from other sites will be done by the clinical nursing coordinator at the primary trial site – University of Ottawa Heart Institute. This will involve regular follow up phone conversations, as well as on site trial visits.
A Data Safety Monitoring Board (DSMB) will be assembled to assess the ongoing conduct of the trial. The DSMB will have at least 3 members with sufficient expertise in aortic surgery, clinical research methods, and statistics. The DSMB will meet twice per year, most often by teleconference, and will provide a summary report to the study team, who will in turn submit it to the Research Ethics Board (REB) and Ottawa Heart Institute Research Corporation (OHIRC) research administration before the due date. The terms of reference for the DSMB have been drafted following OHIRC’s template, and will be stored with the study regulatory files.