Mean arterial pressure is associated with the neurological function in patients who survived after cardiopulmonary resuscitation: A retrospective cohort study

Abstract Background About 18% to 40% of the survivors have moderate to severe neurological dysfunction. At present, studies on mean arterial pressure (MAP) and neurological function of patients survived after cardiopulmonary resuscitation (CPR) are limited and conflicted. Hypothesis The higher the MAP of the patient who survived after CPR, the better the neurological function. Method A retrospective cohort study was conducted to detect the relationship between MAP and the neurological function of patients who survived after CPR by univariate analysis, multivariate regression analysis, and subgroup analysis. Results From January 2007 to December 2015, a total of 290 cases met the inclusion criteria and were enrolled in this study. The univariate analysis showed that MAP was associated with the neurological function of patients who survived after CPR; its OR value was 1.03 (1.01, 1.04). The multi‐factor regression analysis also showed that MAP was associated with the neurological function of patients survived after CPR in the four models, the adjusted OR value of the four models were 1.021 (1.008, 1.035); 1.028 (1.013, 1.043); 1.027 (1.012, 1.043); and 1.029 (1.014, 1.044), respectively. The subgroups analyses showed that when 65 mm Hg ≤ MAP<100 mm Hg and when patients with targeted temperature management or without extracorporeal membrane oxygenation, with the increase of MAP, the better neurological function of patients survived after CPR. Conclusion This study found that the higher MAP, the better the neurological function of patients who survived after CPR. At the same time, the maintenance of MAP at 65 to 100 mm Hg would improve the neurological function of patients who survived after CPR.


| BACKGROUND
Cardiac arrest refers to sudden cardiac arrest accompanied by hemodynamic failure, which leads to severe ischemia, hypoxia, and even death of essential organs (such as the brain). 1 Although significant progress had achieved on the treatment of cardiac diseases and cardiac arrest, the survival rate of them was still only about 10% to 25%. 2,3 About 18% to 40% of the survivors of cardiac arrest have moderate to severe neurological dysfunction, preventing a return to work and normal daily activities. 4 The neurological function of patients with cardiac arrest is related to the sensitivity to ischemic hypoxic brain injury, as well as the duration of cardiac arrest, the quality of cardiopulmonary resuscitation (CPR), the underlying coexisting disease, and post-resuscitation cerebral tissue perfusion. 5 Sekhon et alstudy showed that ischemia and hypoxia were very frequent in patients after CPR, and mean arterial pressure (MAP) was firmly related to oxygenation of brain tissue. 6 At present, studies had shown that the higher the MAP, the less brain damage and the lower the mortality in patients who survived after CPR. 7,8 At the same time, Jakkula et al study shown that MAP was not associated with the markers of brain tissue damage (neuron-specific enolase [NSE] and S100B protein). 9 So, the research about MAP and neurological function of patients survived after CPR is limited and conflicted. In this study, we hope to furtherly explore the relationship between MAP and the neurological function of patients who survived after CPR.

| Study design
A Retrospective cohort study.

| Objective
To explore the relationship between MAP and the neurological outcome of patients who survived after CPR.

| Inclusion criteria
Patients who survived after CPR and admission to intensive care unit (ICU).

| Exclusion criteria
(a) Patients with previous neurological impairment and (b) Patients with the missing value of MAP or 3-month neurological function score.

| Participants
From January 2007 to December 2015, a total of 347 patients survived after CPR and admitted to ICU. Fifty-seven patients were excluded, including 52 patients with previous neurological impairment, four cases with MAP deletion, and once case with 3-month neurological function score deletion. Finally, 290 patients met the inclusion and exclusion criteria.

| Post-resuscitation care
The protocol of post-resuscitation care has been extensively described elsewhere, widely accepted, and applied. [12][13][14]   13. At the same time, to make the model results more reliable, we adopted two ways to select the adjusted variables for multi-factor regression analysis, respectively: (a) According to the clinical significance and P-value <.1 (SOFA score and APACHE II score included MAP, so SOFA and APACHE II score were not included in multi-factor regression analysis to avoid collinearity). Finally, age, diabetes, COPD, out of hospital cardiac arrest, non-cardiac epidemiology, non-shockable rhythm, bystander CPR, witnessed arrest, total epinephrine dose, time to ROSC, acute kidney injury (AKI), lac, and TTM were adjusted in multi-factor regression analysis. (b) If a variable was added to the model, which caused the change of the effective value to be more than 10%, the variable would be adjusted in multi-factor regression analysis.

| Clinical and biochemical data collection
Finally, bystander CPR, total epinephrine dose, and non-shockable rhythm were selected for multi-factor regression analysis. Subgroups analysis were performed, according to MAP, shock, hypertension, SOFA score, ECMO, and TTM, respectively. Only bystander CPR, total epinephrine dose, and non-shockable rhythm were adjusted to avoid overfitting of the model in subgroup analysis.

| DISCUSSION
No matter univariate analysis, multi-factor regression analysis, or multivariate subgroup analysis, all showed that MAP was associated with the neurological outcome of patients who survived after CPR, with the increase of MAP, the neurological outcome of patients also improved.
Brain tissue is susceptible to ischemia and hypoxia. When cardiac arrest occurs, consciousness disorder may occur within 5 to 10 seconds, pupil dilation may occur within 30 to 60 seconds, and irreversible brain injury may occur after cardiac arrest exceeds 4 to 6 minutes. 15 Cerebral blood flow depends on cerebral perfusion pressure, which is equal to MAP minus intracranial pressure. Therefore, MAP is one of the essential factors affecting cerebral blood flow. 16 Govindan et al measured the hemoglobin in brain tissue by using the external spectrum technique, and found that the circulating hemoglobin in brain tissue was correlated with MAP, and also positively correlated with brain tissue damage. 17 According to Hirose et al study, when the MAP dropped during the cesarean section, the regional cerebral blood volume and oxygenation also dramatically declined. 18 This study found that the higher the MAP, the better the neurological outcome of patients who survived after CPR. The possible reasons were as follows: (a). the cerebral vascular autonomic regulation function would be lost due to ischemia and hypoxia of brain tissue after CPR. 19  Hypoperfusion would further lead to ischemia and hypoxia of brain tissue and increase cerebral edema, which would also lead to hypoperfusion of brain tissue, and eventually lead to the aggravation of brain injury and the long-term deterioration of nerve function. 20 (b) When 100 mm Hg ≤ MAP, it would lead to congestive reperfusion, aggravating cerebral edema, and reperfusion injury, due to impaired cerebrovascular autonomic regulation after cardiac arrest. [20][21][22] According to the 2019 AHA guidelines for cardiopulmonary found that MAP was associated with NSE and S100B protein (NSE and S100B were often used to evaluate the degree of brain injury). 27 Ameloot et al study included 112 out-of-hospital cardiac arrest patients, found that higher than reduces brain ischemia, and improves outcome. 8 John Bro-Jeppesen et al study showed that a low MAP was associated with increased mortality. 7 As compared with these studies, our study had the following characteristics: (a). Our study not only included patients with out-of-hospital cardiac arrest, but also patients with in-hospital cardiac arrest, so the findings of our research might be more adaptive. (b) The sample size of our study was relatively more massive, so the credibility of our conclusions might be more substantial. (c) In our study, the variables that might affect the prognosis of cardiac arrest were adjusted to make our findings more

| Limitations of the study
Our study was a retrospective study, so a prospective study should be conducted to verify the findings of our research further. The MAP of our study obtained at ICU admission, it is necessary to investigate the relationship between MAP dynamics and the neurological function of patients who survived after CPR, which would be with more clinical value. Since our study belongs to the reuse of data, the myocardial infarction, emergent catheterization, cardiac etiology, and all-death in our study were missing in the original data, which may be with potential confounders.

| CONCLUSION
This study found that the higher MAP, the better the neurological function of patients who survived after CPR. At the same time, the maintenance of MAP at 65 to 100 mm Hg would improve the neurological function of patients who survived after CPR.