Initial serum cholesterol level as a potential marker for post cardiac arrest patient outcomes

Journal Pre-proof Initial serum cholesterol level as a potential marker for post cardiac arrest patient outcomes Minjung Kathy Chae, Sung Eun Lee, Young Gi Min, Eun Jung Park

PII:

S0300-9572(19)30686-0

DOI:

https://doi.org/10.1016/j.resuscitation.2019.11.003

Reference:

RESUS 8285

To appear in:

Resuscitation

Received Date:

31 July 2019

Revised Date:

30 October 2019

Accepted Date:

1 November 2019

Please cite this article as: Chae MK, Lee SE, Min YG, Jung Park E, Initial serum cholesterol level as a potential marker for post cardiac arrest patient outcomes, Resuscitation (2019), doi: https://doi.org/10.1016/j.resuscitation.2019.11.003

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Initial serum cholesterol level as a potential marker for post cardiac arrest patient outcomes Minjung Kathy Chae¹, Sung Eun Lee¹ ², Young Gi Min¹, Eun Jung Park¹

¹Department of Emergency Medicine, Ajou University Medical Center, Ajou University School of Medicine, Suwon, Republic of Korea

of

²Department of Neurology, Ajou University Medical Center, Ajou University School of Medicine, Suwon, Republic of Korea

ro

Address for correspondence Eun Jung Park

-p

Department of Emergency Medicine, Ajou University Medical Center, Ajou University School of Medicine, Suwon, Republic of Korea

Telephone number: 82-31-219-7750

lP

E-mail address: [email protected]

re

World cup-ro 250 Ajou University Medical Center, Emergency Medicine

Word count of paper: 2173

ur na

Word count of abstract: 185

Abstract

Aim: Cholesterol and lipoproteins have many roles during systemic inflammation in

Jo

critically ill patients. Many previous studies have reported that low levels of cholesterol are associated with poor outcomes in these patients. The aim of this study was to investigate the association of initial total cholesterol with predicting neurologic outcome of post-cardiac arrest patients.

1

Methods: This was a retrospective observational study of out-of-hospital-cardiac arrest (OHCA) survivors who had serum cholesterol levels at admission. Multivariate regression analysis was performed to investigate total cholesterol and its association with neurologic outcome. Area under receiver operator characteristic curve (AUROC) was assessed and cut off values for predicting good or poor neurologic outcomes were analysed.

of

Results: A total of 355 patients were analysed. Lower total cholesterol was significantly associated with poor neurologic outcome [OR: 0.99 (95% CI: 0.98-0.99), p < 0.01] in the

ro

multivariate analysis. Cholesterol was also useful to screening for poor neurologic outcome

neurologic outcome with a specificity of 100%.

-p

[AUROC: 0.70 (95%CI: 0.63-0.77)]. Patients with cholesterol lower than 71 mg/dL had poor

re

Conclusions: Initial cholesterol level is an easily obtained biomarker that showed association

lP

with neurologic outcomes of post cardiac arrest patients.

ur na

Keywords: Heart Arrest ; Cholesterol; Lipoproteins; Critical Illness

Introduction

Jo

During cardiac arrest, whole body ischemia occurs followed by reperfusion injury

after return of spontaneous circulation (ROSC).1,2 This ongoing ischemia-reperfusion injury persists for days with formation of free radical species, systemic inflammation with cytokine production, disseminated endothelial damage and apoptosis, leading to multi-systemic organ damage defined as post cardiac arrest syndrome (PCAS).1,2 As profound neurologic and 2

systemic injury from PCAS results in high mortality and morbidity,2,3 means to prognosticate the outcomes for these patients are important to provide information for families and guide treatment for physicians.1-3 Cholesterol may play an important role in combating systemic inflammation in many critically ill situations. Cholesterol is a precursor of steroid hormones and constitutes cell

of

membranes. About 90% of circulating cholesterol is incorporated into lipoproteins such as high-density lipoprotein (HDL) and low-density lipoprotein (LDL). Lipoproteins can

ro

neutralize and remove bacterial toxins and endotoxins.4-6 They also have anti-inflammatory

-p

properties.5,7 Decreased total cholesterol and lipoproteins have been found to provide information of severity and prognosis for critically ill patients, especially patients with sepsis.

re

4,5,8

lP

Post-cardiac arrest patients have systemic inflammatory responses similar to those of septic patients9. The aim of the present study was to investigate the association of

ur na

cholesterol with outcome in post out-of-hospital cardiac arrest (OHCA) patients.

Methods

Jo

This retrospective study was approved by the institution review board of our hospital (AJIRB-MED-MDB-19-243). Informed consent was waived due to its retrospective nature.

Study setting and participants 3

This was a retrospective observational study performed in a tertiary care teaching hospital with an annual visit of 87,000 patients in South Korea from January 2012 to June 2018. Patients over 18 years of age with admission for post-cardiac arrest care and targeted temperature management (TTM) after out-of-hospital cardiac arrest (OHCA) were included.

of

Patients who did not have an initial serum cholesterol test within 24 hours were excluded.

ro

Post cardiac arrest care and targeted temperature management

Patients with OHCA were managed by standard guidelines of advanced

-p

cardiopulmonary life support. After ROSC, Patients did not receive TTM if (1) they were able to obey to commands (2) had a traumatic cause of arrest (3) brain hemorrhage (4) emergent operation

re

(5) hemodynamic instability (6) do not resuscitate wishes from the family or (7) by physician

lP

discretion. Target temperature of TTM was selected by the treating physician, ranging from

32 to 36c. Multi-faceted critical care management was done following our TTM protocol

ur na

including sedation and shivering control. TTM was performed using cooling devices with a feedback loop system (Artic Sun Energy Transfer Pads, Medivance Corp, Louisville, CO, USA or Cool guard Alsius Icy Heat Exchange Catheter, Alsius Corporation, Irvine, CA, USA). Total cholesterol was initially done by routine practice for most patients using a

Jo

Roche/Hitachi Cobas c analyzer (IN, USA). The median time of cholesterol sampling was 12 minutes after arriving to the emergency department.

Data collection 4

Demographics such as age and sex, underlying disease, prescription of lipid lowering agent, variables related to cardiac arrest such as witnessed arrest, bystander cardiopulmonary resuscitation (CPR), location of arrest, initial shockable rhythm, cause of arrest, CPR duration, initial sequential organ failure assessment (SOFA) and variables related to outcome were collected from our prospective TTM data registry. SOFA score is used to assess organ dysfunction or failure for critical illness severity evaluation., based on the degree of

of

dysfunction of six organ systems. The SOFA score ranges from 0 to 24, with higher scores

ro

having a higher mortality risk10,11. Laboratory findings were retrieved from electronic medical records. We started to follow up patients by phone call at one month for post cardiac arrest

-p

patients since 2018. We started outpatient follow up for good neurologic outcome patients and phone

re

interview for patients since 2019. For patients before 2018, we used cerebral performance

category (CPC) at discharge if we did not have CPC information at one month. CPC 1 and

lP

CPC 2 were defined as good neurological outcomes while CPC 3 to CPC 5 were defined as

ur na

bad neurological outcomes.

Statistical analysis

Continuous variables were reported as median with interquartile range or mean and

Jo

standard deviation depending on whether data had normal distribution. The Wilcoxon ranksum test or two-tailed t test was used to compare continuous variables. Categorical variables were expressed as frequencies and percentages and comparisons were done by using Chisquare test. Baseline characteristics and cholesterol level by neurologic outcome groups were compared. Multivariate logistic regression was performed by backward elimination to 5

analyse the association between total cholesterol and neurologic outcome while adjusting for confounding variables. Receiver operating characteristic (ROC) curves of cholesterol were analysed for verification the value as one of marker for poor neurologic outcome. Cholesterol cut off values with high specificity of predicting poor neurologic outcome was derived. Cholesterol cut off value for predicting good neurologic outcome with maximum Youden Index, the point on the ROC curve with the highest height above the chance line indicating

of

an optimum cut-off point, was also analysed. Data were analysed using Stata software,

Jo

ur na

lP

re

-p

ro

version 15 (Stata Corp LP, TX, USA).

6

3.Results 3.1. Characteristics of study subjects Of 305 post-cardiac arrest patients who were admitted for post-cardiac arrest care and TTM, patients without initial total cholesterol test were excluded (N=12). As a result, a total of 293 patients were analysed (Fig. 1). At one month or discharge, whichever was

of

sooner, 64 patients had good neurologic outcome and 229 patients had poor neurologic outcome. Table 1 displays baseline characteristics along with underlying diseases, CPR

ro

variables and cholesterol by neurologic outcome. Age and male gender were significantly

-p

different between the two outcome groups. Difference in male gender between the two outcome groups was thought to be caused for different proportions of sex by cause of arrest.

re

There were more males than females with cardiogenic cause of arrest [90 (38.5%) vs. 23 (19.0%), p < 0.01]. However, there were less males than females with asphyxia or hanging

lP

as cause of arrest [32 (13.7%) vs. 38 (31.4%), p < 0.01]. Witnessed arrest, initial shockable rhythm, cardiogenic cause of arrest, CPR time and initial SOFA score were also different

ur na

between the two neurologic outcome groups. Cholesterol was significantly lower in the poor neurologic outcome group [170.0mg/dL (121.5-193) vs. 128.0mg/dL (102-153), p < 0.01]

Jo

(Fig. 2.).

3.2. Total cholesterol and outcome Total cholesterol was significantly associated with neurologic outcome [OR: 0.99

(95% CI: 0.98-0.99), p = 0.01] after adjusting for age, sex, witnessed arrest, initial shockable 7

rhythm, CPR time and initial SOFA score (Table 2). Moreover, Initial total cholesterol had a moderate association with poor neurologic outcome [AUROC: 0.69 (95%CI: 0.61-0.77) ] (Fig. 3). Patients with cholesterol lower than 71 mg/dL had poor neurologic outcome. This cutoff had a specificity of 100%. A cut off level of cholesterol ≥ 169.5 mg/dL was able to predict good neurologic outcome with a specificity of 83.4%, a sensitivity of 46.5%, a

of

positive predictive value of 46.5% and negative predictive value of 86.0% (Table 3).

ro

Discussion

Our study was the first study to seek association and cutoff levels of total cholesterol

-p

in screening outcomes of post-cardiac arrest patients. Low level of serum total cholesterol at

re

admission was associated with poor neurological outcome. The association of total cholesterol level with poor outcome was moderate. Patients with total cholesterol level lower

lP

than 71 mg/dL had poor outome with specificity of 100%. Total cholesterol level at admission is an easily obtained biomarker that was associated with outome in PCAS patients.

ur na

There are some possible causes for the association between low cholesterol level and poor outcome patients after cardiac arrest. First, lipoprotein consumption from neutralizing abundant endotoxins can occur and cholesterol levels can be decreased by the severity of inflammation after arrest.9,12,13 In PCAS patients, systemic inflammation continues with

Jo

increment of proinflammatory cytokine and endotoxemia. The degree of endotoxemia is associated with increased severity score and hemodynamic instability.9,12 Cholesterol and lipoproteins have many critical roles during systemic inflammation in various critical situations.13 Lipoproteins are involved in removal of endotoxins by binding to 8

lipopolysaccharide and lipoteicholic acid to neutralize the inflammatory process, leading to monocyte and cytokine reduction.7,8,13-16 Lipoproteins can also reduce thrombotic properties resulting from inflammation by inhibiting endothelial adhesion molecules.17,18 Multiple studies have shown that low total cholesterol level is detected in various systemic inflammatory critically ill patients, especially in patients with poor outcome.5,7,8,19,20 In a previous study of patients with sepsis, total cholesterol was a better prognostic predictor than

of

procalcitonin or C-reactive protein for survival.16 In addition, patients with cholesterol levels

ro

lower than 50mg/dL had 82% mortality.16 In another study of patients with prolonged sepsis, total cholesterol was the most predictive factor for mortality among 15 biomarkers

-p

investigated.8 Second, low cholesterol may be caused by impaired synthesis and secretion

re

due to liver dysfunction after arrest. Nagase et al. have investigated time course changes in total cholesterol, plasma cholesterol ester (CE) and free cholesterol (FC) in PCAS patients.6

lP

In patients with early mortality, total cholesterol was consistently low while FC/CE ratio was high, implying that liver dysfunction was a key determinant in this process.6,16 Third, whether

ur na

hypocholesterolemia was a result of critical illness and secondary manifestation/indicator of the severity of disease or a pre-existing risk factor for combating systemic inflammation even before cardiac arrest is unclear. Whether cholesterol depletion actively contributed to the ongoing deterioration and resulted in poor neurologic outcome during PCAS is also unknown.

Jo

Healthy individuals with lower cholesterol have been found to have fewer lymphocytes, fewer total T cells and fewer CD8+ cells compared to individuals with higher cholesterol.21 Furthermore, when Lagrost et al. have investigated cholesterol levels before, during and after cardiac surgery and cardiopulmonary bypass, they found that low cholesterol level even 9

before surgery was associated with a high risk of sepsis after surgery19. These findings imply that hypocholesterolemia could be a risk factor before the critical event. Cholesterol is also a component of cell membrane and a principal component for steroid synthesis. Hypocholesterolemia may cause relative adrenal insufficiency and shock in post cardiac arrest patients as in other critical illnesses.22 Fourth, the association of cholesterol and neurologic outcome could be affected by other factors.23,24 Old age is associated with lower

of

cholesterol levels and poor neurologic outcome.24 Also, initial shockable rhythms or

ro

cardigenic causes of arrest could be associated with high cholesterol levels and good neurologic outcome.23 From our data, when we performed a univariate logistic regression of

-p

poor neurologic outcome and cholesterol by presumed cardiogenic arrest, cholesterol was

re

not associated with outcomes in non-cardiogenic arrest (OR 0.99 95%CI 0.98-1.00), whereas it was associated with outcomes in cardiogenic arrest (OR 0.98. 95%CI 0.97-0.99). However,

lP

in the univariate logistic regression of poor neurologic outcome and cholesterol by initial shockable rhythms, cholesterol was associated with outcomes in both non-shockable rhythms (OR

ur na

0.99 95%CI 0.98-0.99) and shockable rhythms (OR 0.98 95%CI 0.97-0.99). In the multivariate analysis we used initial shockable rhythm because of the objectiveness. Although, cholesterol was

associated with neurologic outcome with these variables adjusted in the multivariate logistic regression model, differences in age or cause of arrest could have affected our analysis.

Jo

Further studies are needed to investigate the pathophysiologic mechanisms to know if reversing the depletion of cholesterol could lead to potential therapeutic options in the future for PCAS patients.25,26 Two previous studies have investigated lipid profiles in post cardiac arrest patients. 10

In one study with 59 PCAS patients, high HDL and apolipoprotein A1 was associated with good neurological outcome.13 Total cholesterol level was lower, but not significantly lower in poor neurologic outcome patients in this study possibly due to the limited number of patients.13 The other study on cholesterol metabolism in post cardiac arrest patients by Nagase et al. investigated total cholesterol and lipoprotein secretion deficiency by liver function. For patients with early mortality, total cholesterol, FC and CE decreased with high

of

FC/CE ratio, whereas these levels were higher and unchanged throughout the course of

ro

treatment in patients who survived.6 These findings coincide with findings of our study. The present study has some limitations. First, this is a retrospective single center

-p

study. Although a single center study is a setting that can provide fairly homogenious

re

treatment to patients, further studies may be needed before generalizing results of this study. Second, as described earlier, the exact mechanism or results of our study cannot be

lP

determined. How much low cholesterol contributed to poor neurologic outcome as a preexisting risk factor and how much it was responsible for the consequence of the severity of

ur na

arrest and acute phase response were not clear. Third, serum cholesterol may not be able to decrease due to impaired cholesterol excretion in bile in cholestasis.27 We did correct for SOFA score that included total bilirubin level in the scoring system in our multivariate analysis with consideration of this. However, other unknown confounding factors might have

Jo

affected findings of our study. Fourth, some patients that visited our hospital before 2018 did not have one month CPC information. There were 67 patients that had CPC addressed at discharge. Among these patients, 59 patients had a discharge CPC of 4. In Korea, we were not able to perform withdrawal of life support until February, 2018. Even since February 2018, withdrawal of life support is performed rarely in a very conservative mannor. Most of our severely brain injured patients that 11

remain unconscious are transferred out for long term supportive care because we did not withdraw life support. Although we did not have one month CPC for some patients before 2018, as most of them were CPC 4 with severe brain injury we do not think that one month CPC would not be very different from the discharge CPC in these patients. Fifth, this study analysed initial total

cholesterol levels of PCAS patients. Further information of components of cholesterol and lipid profile as well as serial measurements of cholesterol during the dynamic phase of post

ro

of

cardiac arrest will be needed for better understanding of our findings.

4. Conclusions

-p

In conclusion, the present study showed that low total cholesterol level at admission was associated with poor neurologic outcome in PCAS patients. Cholesterol is an easily

re

obtained biomarker whose association with neurologic outcome after OHCA merits further

ur na

Conflicts of Interest

lP

study.

The authors have no conflict of interest to declare.

Jo

Conflict of interest

The authors have no commercial associations or sources of support that might pose a conflict of interest.

Acknowledgements 12

This study was supported by the faculty research fund of Ajou University School of Medicine but has no conflicts of interest. (M.C and E.P)

Jo

ur na

lP

re

-p

ro

of

M.C. and S.L. have a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. 2018R1C1B60035) but has no conflicts of interest.

13

References 1.

Nolan JP, Neumar RW, Adrie C, et al. Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication. A Scientific Statement from the International Liaison Committee on Resuscitation; the American Heart Association Emergency Cardiovascular Care Committee; the Council on Cardiovascular Surgery and Anesthesia; the Council on Cardiopulmonary, Perioperative, and Critical Care; the Council on Clinical Cardiology; the Council on Stroke. Resuscitation 2008;79:350-79. Mongardon N, Dumas F, Ricome S, et al. Postcardiac arrest syndrome: from immediate resuscitation

of

2.

3.

ro

to long-term outcome. Ann Intensive Care 2011;1:45.

Nolan JP, Soar J, Cariou A, et al. European Resuscitation Council and European Society of Intensive

-p

Care Medicine Guidelines for Post-resuscitation Care 2015: Section 5 of the European Resuscitation Council Guidelines for Resuscitation 2015. Resuscitation 2015;95:202-22. Cirstea M, Walley KR, Russell JA, Brunham LR, Genga KR, Boyd JH. Decreased high-density

re

4.

lipoprotein cholesterol level is an early prognostic marker for organ dysfunction and death in patients

5.

lP

with suspected sepsis. J Crit Care 2017;38:289-94.

Lekkou A, Mouzaki A, Siagris D, Ravani I, Gogos CA. Serum lipid profile, cytokine production, and clinical outcome in patients with severe sepsis. J Crit Care 2014;29:723-7. Nagase M, Sakurai A, Sugita A, et al. Oxidative stress and abnormal cholesterol metabolism in patients

ur na

6.

with post-cardiac arrest syndrome. J Clin Biochem Nutr 2017;61:108-17.

7.

Guirgis FW, Donnelly JP, Dodani S, et al. Cholesterol levels and long-term rates of communityacquired sepsis. Crit Care 2016;20:408. Yamano S, Shimizu K, Ogura H, et al. Low total cholesterol and high total bilirubin are associated

Jo

8.

with prognosis in patients with prolonged sepsis. J Crit Care 2016;31:36-40.

9.

Adrie C, Adib-Conquy M, Laurent I, et al. Successful cardiopulmonary resuscitation after cardiac arrest as a "sepsis-like" syndrome. Circulation 2002;106:562-8.

10.

Cour M, Bresson D, Hernu R, Argaud L. SOFA score to assess the severity of the post-cardiac arrest syndrome. Resuscitation 2016;102:110-5. 14

11.

Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22:707-10.

12.

Grimaldi D, Sauneuf B, Guivarch E, et al. High Level of Endotoxemia Following Out-of-Hospital Cardiac Arrest Is Associated With Severity and Duration of Postcardiac Arrest Shock. Crit Care Med 2015;43:2597-604.

13.

Son YS, Kim KS, Suh GJ, et al. Admission levels of high-density lipoprotein and apolipoprotein A-1

of

are associated with the neurologic outcome in patients with out-of-hospital cardiac arrest. Clin Exp Emerg Med 2017;4:232-7.

Flegel WA, Baumstark MW, Weinstock C, Berg A, Northoff H. Prevention of endotoxin-induced

ro

14.

monokine release by human low- and high-density lipoproteins and by apolipoprotein A-I. Infect

15.

-p

Immun 1993;61:5140-6.

Berbee JF, Havekes LM, Rensen PC. Apolipoproteins modulate the inflammatory response to

Biller K, Fae P, Germann R, Drexel H, Walli AK, Fraunberger P. Cholesterol rather than procalcitonin

lP

16.

re

lipopolysaccharide. J Endotoxin Res 2005;11:97-103.

or C-reactive protein predicts mortality in patients with infection. Shock 2014;42:129-32. 17.

Cockerill GW, Saklatvala J, Ridley SH, et al. High-density lipoproteins differentially modulate

ur na

cytokine-induced expression of E-selectin and cyclooxygenase-2. Arterioscler Thromb Vasc Biol 1999;19:910-7.

18.

Montero-Chacon LB, Padilla-Cuadra JI, Chiou SH, Torrealba-Acosta G. High-density lipoprotein, mean platelet volume, and uric acid as biomarkers for outcomes in patients with sepsis: an

Jo

observational study. J Intensive Care Med 2018:885066618772825.

19.

Lagrost L, Girard C, Grosjean S, et al. Low preoperative cholesterol level is a risk factor of sepsis and poor clinical outcome in patients undergoing cardiac surgery with cardiopulmonary bypass. Crit Care Med 2014;42:1065-73.

20.

Gordon BR, Parker TS, Levine DM, et al. Relationship of hypolipidemia to cytokine concentrations and outcomes in critically ill surgical patients. Crit Care Med 2001;29:1563-8. 15

21.

Muldoon MF, Marsland A, Flory JD, Rabin BS, Whiteside TL, Manuck SB. Immune system differences in men with hypo- or hypercholesterolemia. Clin Immunol Immunopathol 1997;84:145-9.

22.

Marik PE. Critical illness-related corticosteroid insufficiency. Chest 2009;135:181-93.

23.

Nagasawa SY, Okamura T, Iso H, et al. Relation between serum total cholesterol level and cardiovascular disease stratified by sex and age group: a pooled analysis of 65 594 individuals from 10 cohort studies in Japan. J Am Heart Assoc 2012;1:e001974.

24.

Yi SW, Yi JJ, Ohrr H. Total cholesterol and all-cause mortality by sex and age: a prospective cohort

25.

Morin EE, Guo L, Schwendeman A, Li XA. HDL in sepsis - risk factor and therapeutic approach.

ro

Front Pharmacol 2015;6:244. 26.

of

study among 12.8 million adults. Sci Rep 2019;9:1596.

Kwon WY, Suh GJ, Kim KS, Kwak YH, Kim K. 4F, apolipoprotein AI mimetic peptide, attenuates

-p

acute lung injury and improves survival in endotoxemic rats. J Trauma Acute Care Surg 2012;72:157683.

Chiarla C, Giovannini I, Giuliante F, et al. Severe hypocholesterolemia in surgical patients, sepsis, and

re

27.

Jo

ur na

lP

critical illness. J Crit Care 2010;25:361.e7-.e12.

16

of ro -p

Jo

ur na

lP

re

Figure 1

Figure 2

17

of ro -p

Figure 3

re

Figure 1. Study Enrollment. Figure 2. Cholesterol level by neurologic outcome

total cholesterol.

lP

Figure 3. Receiver operator characteristic curve for prediction of poor outcome using

Jo

ur na

AUROC (95% Confidence interval) = 0.69 (0.61-0.77)

18

Poor outcome (n=229)

P value

52.0 (14.0)

57.9 (18.8)

0.02

Male gender (%)

50 (78.1)

141 (61.6)

0.01

Hypertension

24 (37.5)

80 (34.9)

0.62

Diabetes

16 (25.0)

57 (24.9)

0.65

Lipid lowering agent prescription

10 (15.6)

18 (7.9)

0.06

Witnessed arrest (%)

56 (87.5)

143 (62.5)

<0.01

Bystander CPR (%)

43 (67.2)

140 (61.1)

0.38

41 (17.9)

<0.01

re

-p

Age (years)

50 (78.1)

lP

Initial shockable rhythm (%) Arrest location (%) Home Public Healthcare facility Other Unknown

ro

Good outcome (n=64)

of

Table 1. Baseline characteristics of study population

107 (46.7) 52 (22.7) 42 (18.3) 20 (8.7) 8 (3.5)

Cardiogenic cause of arrest (%)

50 (78.1)

50 (21.8)

<0.01

CPR time (min)

16 (11-26)

26 (18-40)

<0.01

Initial SOFA score

8.5 (6-10)

10 (8-13)

<0.01

128 (102-153)

<0.01

Jo

ur na

21 (32.8) 26 (40.6) 8 (12.5) 7 (10.9) 2 (3.1)

Cholesterol (mg/dl)

170 (121.5-193)

Data are presented as mean (SD) , median(IQR) or n(%). CPR, cardiopulmonary resuscitation; SOFA, Sequential organ failure assessment.

19

0.05

Table 2. Multivariate logistic regression of total cholesterol and neurologic outcomes

95% Confidence Interval

Age

1.02

(1.00-1.05)

0.05

Male gender

0.58

(0.24-1.43)

Cholesterol (mg/dl)

0.99

(0.98-0.99)

0.01

Witnessed arrest

0.25

Initial shockable rhythm CPR time (min)

ro

0.01

0.08

(0.04-0.19)

<0.01

1.06

(1.02-1.09)

<0.01

1.10

(0.96-1.25)

0.18

re

-p

(0.09-0.71)

CPR,cardiopulmonary resuscitation; SOFA, Sequential organ failure assessment.

Jo

ur na

Pseudo R² 0.42

0.24

lP

Initial SOFA score

P value

of

Odds ratio

20

Table 3. Cutoff and diagnostic value of cholesterol for predicting neurologic outcome Cut off* Cholesterol predicting poor outcome

Cholesterol predicting good outcome

<71 ≧169.5

Sensitivity Specificity

PPV

NPV

7.0 %

100.0 %

100.0% 23.1%

51.6 %

83.4 %

46.5 % 86.0 %

of

The cut off value to predict poor neurologic outcome was set to maximize specificity in predicting poor neurologic outcome, and the cut off value of cholesterol to predict good neurologic outcome is the cut off

ro

value of maximum Youden index(0.35). * Cholesterol level (mg/dl); PPV, positive predictive value; NPV,

Jo

ur na

lP

re

-p

negative predictive value.

21