The Physiology of Circulatory Arrest in Humans

MAIN STUDY

PURPOSE The goal of this project is to characterize the cerebrovascular physiology of circulatory death during withdrawal of life sustaining measures in humans.

JUSTIFICATION Death is defined as the permanent loss of brain function following absence of cerebral blood flow (CBF). Such physiology can result from two clinical scenarios: a) absent CBF in patients with a beating heart (neurological brain death) or b) after natural circulatory arrest. In the latter scenario, eventual cessation of innate cardiac output leads to absent CBF, brain tissue oxygenation and irreversible loss of brain function. Pertinent to solid organ transplantation, death must occur prior to consideration of organ donation. In the setting of circulatory death, withdrawal of life-sustaining measures (WLSM) is undertaken and organ donation commences after the declaration of death. This process is termed controlled donation after cardiac death (DCD). Due to the scarcity of available organ donation opportunities, the emergence of DCD has provided increased access to life-saving solid organ transplants for recipients. However, challenges remain with respect to successes of DCD graft organ function and optimal recipient outcomes. Principally, the timeliness from WLSM to circulatory death declaration is crucial to graft viability with prolonged periods associated with worse graft organ ischemia, function and adverse recipient outcomes. To this extent, the determination of the precise timing of death determination during DCD is imperative to inform timely diagnosis of death and optimize retrieval of high-quality grafts for solid organ transplant recipients.

The characterization of the cerebrovascular physiology of circulatory death in humans has knowledge gaps. Current guidelines define a pulse pressure < 5mmHg (systolic - diastolic blood pressure) using radial arterial line monitoring as the acceptable threshold for timing of circulatory death. Although pragmatic, this definition assumes that in vivo hemodynamic physiology generated by the heart (reflected by the arterial line) is simultaneously reflected in the cerebral circulation. It is well known that systemic hemodynamics do not necessarily correlate with cerebral hemodynamics during critical illness. Pathophysiologic sequelae (e.g. elevated intracranial pressure) in critically ill acutely brain injured patients who undergo WLSM suggest that cerebral perfusion may cease before systemic hemodynamics (detected by radial arterial line monitoring) deteriorate irreversibly. As such, it is plausible that cerebral perfusion ceases before a pulse pressure < 5mmHg is achieved thereby setting the conditions by which the physiologic requirement of circulatory death is present (absent CBF) but the clinical definition has not (pulse pressure < 5mmHg) been reached. Clinically, this scenario would expose viable graft organs to prolonged ischemia and negatively affect solid organ transplant recipient outcomes

HYPOTHESES AND AIMS

OBJECTIVES and associated HYPOTHESIS Aim 1 (Cerebral Hemodynamics): To determine the pulse pressure (measured with an in situ radial arterial catheter) at which cerebral hemodynamics cease (measured with transcranial Doppler derived middle [MCA-Fv] and posterior cerebral artery blood flow velocities [PCA-Fv]) following WLSM. The investigators hypothesize that cerebral hemodynamics will cease when the pulse pressure is > 5mmHg.

Aim 2 (Feasibility): To measure the feasibility to conducting multimodal monitoring and data collection during the dying process in humans until circulatory arrest. Specific physiologic variables being collected during the dying process include telemetry, peripheral oximetry, mean arterial pressure monitoring, PCA-Fv \& MCA-Fv, jugular venous bulb oximetry, pulmonary artery mixed venous oximetry and regional oxygen saturation (from near infrared spectroscopy). The investigators hypothesize that feasibility will be strong and we will successfully gather > 80% of data points for each case.

Aim 3) To evaluate relationship between mean arterial pressure (mmHg) and brain blood flow velocity (cm/sec) during the dying process in humans until circulatory arrest.

Aim 4) To evaluate relationship between mean arterial pressure (mmHg) and oxygen extraction fraction (%) and regional oxygen saturation (%) during the dying process in humans until circulatory arrest.

Aim 5: The investigate the impact of brain biomarker release during the ischemic phase of the dying process in humans until circulatory arrest. The investigators hypothesize that brain biomarker levels will increase during the dying process.

Aim 6 (Pathology): To conduct post-mortem heart and brain autopsy to contextualize and associate tissue-based injury characteristics with physiologic observations. The investigate hypothesize that brain pathology will be related to physiologic observations.

RESEARCH DESIGN This is a prospective observational study of the evaluation of the cerebrovascular physiology of circulatory death.

STATISTICAL ANALYSIS Our sample size is one of convenience. We will enroll 50 participants. Discrete variables will be summarized by frequencies and percentages. Continuous variables will be summarized by mean (SD) or median (IQR) if data are skewed.

Aim 1: The pulse pressure at which MCA-Fv and PCA-Fv cease will be summarized as a median and range across the cohort. The percentage of patients exhibiting absent CBF with a pulse pressure of > 5mmHg will be stated as well. Sub Aim 1b: Random slope and intercepted linear mixed effects models will be used to assess relationships between cerebrovascular and cardiovascular variables during the dying process in humans until circulatory arrest.

Aim 2: The feasibility of data elements collected will be reported by percentage of data elements obtained in each case across the cohort. In each case, a dichotomous (yes/no) reported collection of the data element will be reported. The overall percentage of data elements collected will be reported for each case and across the entire cohort as a numerical measure of feasibility.

Aim 3: Relationships between select variables will be analyzed with random slope, random intercept linear mixed-effects model with continuous predictors (MAP,) as fixed factors and patient as a random factor. Cerebrovascular responses to progressive hypotension (MAP) following WLSM will be determined via random slope and random intercept linear mixed effects models with MAP or pulse pressure as a continuous predictor and patient as a random factor.

Aim 4: Wilcoxon signed-rank tests will be used to evaluate the change in brain biomarker levels during the dying process in humans until circulatory arrest.

Aim 5: Kruskal-Wallis and Dunn's post-hoc correction will be used to assess the relationship between brain pathology and continuous physiologic variables.

Sex Based Analysis: Assessing sex-related differences will encompass an exploratory analysis to inform future hypothesis generation. In our neuromonitoring patients, the proportion of patients has been approximately 60% male and 40% female. The investigators expect a similar ratio of males to females in the current study. Data will be stratified by sex, across all patients and within each study group. Further, for both study groups the investigators will assess differences between males and females by including an interaction term group*sex into our linear mixed models.

PATHOLOGY SUB STUDY

PURPOSE To perform post-mortem pathological examination of nervous system tissue (brain and spinal cord)

HYPOTHESIS The investigators hypothesize that the brain will show histological features correlated with early cessation of blood perfusion such as abnormalities in and around small vessels and in the blood-brain barrier, tissue edema, and markers of early cellular necrosis, and that these changes will not be present, or will be present to a much lesser extent, in the spinal cord of the same individuals.

JUSTIFICATION AND OBJECTIVES Post-mortem brain and spinal cord autopsy will be obtained in patients to provide a tissue correlate and context to the neuromonitoring data. The investigators will evaluate whether two distinct areas of the central nervous system (brain and spinal cord) with different blood supply show the same degree of histopathological changes. Specifically, pathologic examination will be undertaken to provide macroscopic evaluation of the middle and posterior cerebral arteries for signs of vascular pathology (e.g. intravascular thrombi, vasoconstricted state) to provide mechanistic explanations for the anticipated results. Examination will also be undertaken in the regions of anterior and posterior circulation to assess microvascular pathology that could explain the cerebrovascular physiologic results. This will also be compared to the spinal cord which has a different blood supply from the brain. Specifically, the formalin fixed samples will be used for hematoxylin and eosin and special stains to assess for abnormalities in and around small vessels and in the blood-brain barrier, tissue edema, and markers of early cellular necrosis.

RESEARCH DESIGN This single-center prospective post-mortem pathological analysis of patients who underwent WLST will describe the pathologic characteristics of neurological injury in the neurovascular unit.

STATISTICAL ANALYSIS The pathology tissue analysis will be used to draw descriptive analyses pertaining to the macrovascular and microvascular abnormalities present on post-mortem examination. The change in the concentrations of the arterial and cerebral arterio-venous gradients of the brain biomarkers will be analyzed before WLST and immediately following circulatory death using a one-way ANOVA analysis. Comparisons between the pulse pressure at which cerebral hemodynamics, brain oxygenation and neural electrical activity cease will be conducted between patients with an antecedent acquired brain injury versus those without using a Kruskal-Wallis test.