a b s t r a c t

Purpose: The aim of our study is to research the role and efficacy of cerebral oximetry in predicting neurologic prognosis when applied during TTM to patients experiencing coma after CA. Methods: This study was performed on surviving adult Comatose patients after CA treated with TTM. The average scores of rSO₂ was measured at 6 h intervals for the first 2 days and once a day for the following 3 days with a NIRS device during TTM. The CPC scale was used to define the neurologic outcomes of patients. We compared the correlations of rSO₂ values between good (CPC 1-2) and poor (CPC 3-5) neurologic outcomes in CA patients. Results: There was no statistically significant difference identified between the prognosis groups in terms of rSO₂, CPR durations, hemoglobin values and admission body temperature (p N 0.05). When the variation in rSO₂ values over time is investigated, though there was no significant difference between the good and poor prognosis groups, it appeared to fall in the first 6 h in both prognosis groups. The median NT-proBNP and lactate values were observed to be higher in the poor prognosis group.

Conclusion: There is no significant correlation between rSO₂ values and neurologic outcomes. Multimodal mon- itoring methods may be useful and further studies with a larger patient population are necessary in this area.

(C) 2018

1. Introduction

Cardiac arrest, whether inside or outside the hospital, causes high rates of mortality and morbidity. In patients with return of spontaneous circulation (ROSC), there is short or long term high risk of death with hypoxic-ischemic neurologic injury. In the short term it is difficult to de- termine the degree of brain damage and neurologic prognosis [1-3].

The use of Targeted temperature management (TTM), post-cardiac arrest interventions (PCAI) and neurologic prognostication methods are recommended to increase the chances of survival and to improve neurologic outcome [3-6].

Hypothermia is the only treatment method shown to have the abil- ity to increase both survival and to improve neurologic outcomes in pa- tients after CA [7,8]. TTM is widely recommended for comatose patients after cardiac arrest [9].

A variety of neuromonitoring methods may be applied to ensure neurologic protection or to predict neurologic outcomes. Many neuromonitoring models observing Oxygen extraction and consump- tion require invasive interventions, such as jugular venous bulb oxygen saturation (SjvO₂) [10].

* Corresponding author.

E-mail address: [email protected]. (A. Saritas).

Near infrared spectroscopy (NIRS), which can be applied bedside and is non-invasive, provides real time information relating to the re- gional cerebral oxygen saturation (rSO₂) in the frontal lobe [1,4]. There are very few studies on the potential role and reliability of cere- bral oximetry to predict neurologic prognosis in post-CA patients and there is no clear consensus.

The aim of our study is to research the role and efficacy of cerebral oximetry in predicting neurologic prognosis when applied during TTM to comatose patients after CA.

1. Material and methods

This study was approved by the institutional review board and ethics committee of Izmir Tepecik Training and Research Hospital, Faculty of Medical Sciences (16/08/2016, 29/3), and written informed consent was obtained from patient’s next of kin. The clinical trials (clinicaltrials.gov) registration number is NCT03024021.

Study design

This prospective observational study was performed between Sep- tember 2016 to December 2017 on surviving adult comatose patients after CA treated with TTM in the tertiary intensive care of Tepecik

https://doi.org/10.1016/j.ajem.2018.04.016

0735-6757/(C) 2018

Training and Research Hospital. Twenty-five patients who met inclusion criteria were included in the study. The inclusion criteria: N18 years of age, comatose patients after CA. Exclusion criteria: b18 years of age, Glasgow coma scale (GCS) N 8, trauma, intracerebral hemorrhage, major hemorrhage and poor neurologic status before CA.

Procedures

According to tertiary intensive care protocol at our hospital, TTM is applied to comatose patients successfully resuscitated after CA within indications. Those with suspected falls based on clinical findings or an- amnesis, and patients without clear cause of CA identified had com- puted tomography screening applied before Coronary angiography to exclude intracerebral pathologies. Patients with indications had CAG administered as soon as possible, with percutaneous coronary intervention (PCI) applied. After the procedure patients were admitted to the intensive care unit and TTM was initiated immediately.

TTM protocol

All patients were intubated and monitored with mechanical ventila- tion support. In this study no patient had cooling interventions applied before reaching the hospital and all patients had TTM applied after admission to the hospital. Cooling began as soon as possible after admis- sion with cold saline and continued with TTM Surface cooling systems (ArcticGel(TM) pads, Arctic Sun(R) 5000, Medivance, Louisville, CO, USA). Patients were cooled to the target temperature of 35 ?C as quickly as possible. Maintenance of the target temperature was ensured using a TTM device for 24 h. After the 24 h hypothermia period of TTM treat- ment ended, reheating began at a rate of 0.25 ?C/h until 36.5 ?C was reached. The feedback cycle to control the target temperature was pro- vided by an esophageal probe integrated with the TTM device. This nor- mothermic period of TTM treatment continued until the patient’s 5th day of admission.

Patient’s management and sedation

All patients had TTM applied with a standard protocol and patients were sedated for the hypothermia period. Sedation was ensured with intravenous propofol (50 ug/kg/min) and remifentanil (0.1 ug/kg/min) (if tolerated hemodynamically). In situations with muscle shiv- ering during temperature management, firstly magnesium sulfate 0.5-1 mg/h iv (target 3-4 mg/dl) and meperidine 25-50 mg iv were administered. In situations with no response, shivering was prevented by administering rocuronium. All patients had invasive arterial moni- toring. If necessary, noradrenalin and/or inotrope infusion was titrated to ensure mean arterial pressure above 65 mmHg. After rewarming se- dation was reduced based on the patient’s comfort, and then ended. When weaning criteria were met (when neurologic, hemodynamic and respiratory systems had sufficiently improved), patients were extubated.

Data management

Demographic data, cardiopulmonary resuscitation (CPR) durations, whether CAG/PI was applied, presence of Epileptic seizures and admission NT-proBNP (N terminal B type natriuretic peptide) values were recorded. NT-proBNP measurement from CA patients is a routine procedure in our ICU. On admission to intensive care, the patients’ mean arterial pressure (MAP), heart rate (HR), peripheral oxygen satu- ration (SaO₂), rSO₂ values, body temperature and arterial blood gas samples (PaO₂ - PaCO₂) began to be recorded to find if any relation be- tween prognosis groups. Data were recorded on admission to intensive care and at 6 h, 12, h, 18 h, 24 h, 48 h and 3rd, 4th and 5th days. Throughout the study, patients’ Neurologic assessment was recorded by evaluating Glasgow coma scale (GCS), FOUR (full outline of

unresponsiveness) score coma scale and cerebral performance category scale (CPC) except the hypothermia period of TTM due to sedation. At time of discharge from hospital and 90 days after discharge CPC scores were evaluated again.

Cerebral oximetry

In our intensive care unit cerebral tissue oxygen saturation is routinely measured continuously with an NIRS device (Somanetics INVOS, Covidien, Boulder, CO, USA) for post CA patients. Sensors were bilaterally applied to both frontotemporal areas. rSO₂ observation began as soon as possible after admission to intensive care.

Data was collected every 2 s during TTM. The average scores of mean rSO₂ values were measured at 6 h intervals for the first 2 days and once a day for the following 3 days. Physicians and nurses checked the signal quality every 3 h to ensure sufficient monitoring. Data were obtained from the device with a flash disk and checked for analysis.

Neurologic outcome

The correlation between rSO₂ level and the scoring systems and the association of this level with early determination of neurologic prognos- tic outcomes was assessed. The CPC scale was used to define the neuro- logic outcomes of patients. The primary outcome was defined as CPC scale reaching the best level on discharge from hospital. According to the CPC scale: CPC 1 indicates good cerebral performance, CPC 2 implies a moderate disability (sufficient for independent activities in daily life), CPC 3 indicates severe disability (dependent on others), CPC 4 implies coma or vegetative state and CPC 5 stands for death. CPC 1-2 were assessed as Good neurologic outcome (good prognosis group) and CPC 3-5 were assessed as Poor neurologic outcomes (poor prognosis groups). During discharge, the CPC levels and rSO₂ measurements were determined by at least 2 physicians blind to the study. In cases of disagreement, discussions were held and consensus reached. The cor- relation of CPR durations and NT-proBNP, used as biomarker, with rSO₂ values and effects on prognosis estimation were assessed.

Ninety days after discharge, neurologic assessment of patients was completed by a clinician blind to the study. The secondary outcome of mortality rates within 90 days after discharge was assessed.

Statistical analysis

Parametric analyses were applied to variables showing normal dis- tribution, while non-parametric analyses were applied to variables without normal distribution. For paired group comparisons the inde- pendent samples t-test/Mann Whitney U tests were used to evaluate the groups. Repeated measure ANOVA was performed for evaluate the time and groups affects. For comparison of categorical data the chi square analysis was used. For prediction of prognosis the ROC (receiver operating characteristic) analysis was used to calculate sensitivity and specificity values.

1. Results
   1. General characteristics and mortality

During the study period, 32 patients with successful ROSC were ad- mitted to the intensive care unit. Seven patients were excluded from the study; 3 due to intracerebral hemorrhage, 2 were died in the first 18 h and 2 previously had poor neurologic status.

The research included a total of 25 individuals, 19 male (76%) and 6 female (24%). Though 2 patients were exitus 1 day before the end of the normothermic period of TTM and another 2 patients were died 2 days before the end of this period, they were included in the study due to completing 72 h of TTM (hypothermia and normothermic periods).

Table 1

Demographic datas and mortality statistics.

Table 3

Comparison of prognostic groups with regional cerebral oxygen saturation, CPR duration, body temperature and NT-proBNP.

Patients Survivors Nonsurvivors All groups p

Sex

N (%) N (%) N (%)

Variables CPC 1 2

Good Prognosis

_X2 N Med (Min,

CPC 3 4 5 p

Poor Prognosis

N Med (Min,

Male 6 (%31,6) 13 (%68,4) 19 (%76) 0,637

Max)

Max)

Female 1 (% 16,7) 5 (%83,3) 6 (%24)

GCS 7 6,6 (3, 10) 18 3 (3,8) 0,012^MW

CAG/PCI

Performed 5 (%71,4) 7 (%38,9) 12 (%48) 0,202^X2

Not performed 2(%28,6) 11 (%61,1) 13 (%52)

Epileptic seizure

Present 5 (%71,4) 13 (%72,2) 18 (%72) 0,663^X2

Not present 2(%28,6) 5 (%27,8) 7 (%28)

NT-proBNP 5 122 (15, 479) 15 610 (36,

5000)

N Mean +- Sd N Mean +- Sd

FOUR score	7	8.9 +- 3	18	5,3 +- 2,7
rSO2 admission	7	71,4 +- 12,9	18	69,9 +- 12

*

0,040

MW?

0,011^t ?

0,536^t

Age

Mean +- sd 46,6 +- 18,1 52,4 +- 16,0 50,8 +- 16,5 0,435^t

rSO₂ 5. day 7 66,9 +- 12,1 16 64,2 +- 10,2 0,626^t

CPR duration (min) 7 16,9 +- 6,8 18 18,1 +- 11,4 0,799^t

Hemoglobin 7 12.2 +- 1,39 18 12.7 +- 2.25 0,581^t

X²: Sex, CAG/PCI, epileptic seizure.

Body temperature admission (?C)

7 36,9 +- 0,48 18 37,2 +- 1,11 0,594^t

t-test: age.

CAG: Coroner angiography. PCI: Percutaneous coronary interventions. p b 0.05 anlamlilik duzeyi.

MW: Mann Whitney U test. t: t-test.

Note: Values in bold are statistically significant.

* p b 0,05.

When demographic data were investigated, there was no statististically significant difference in terms of mortality (Table 1).

For 12 patients (48%) PCI was performed while it was not performed

for 13 (52%). When the association with mortality is investigated according to whether PCI was performed or not, there was no statis- tically significant association between died and surviving individuals (p N 0.05).

While 18 patients (72%) had epileptic seizures, 7 (28%) did not. There was no significant correlation between epileptic seizure presence and mortality (p N 0.05) (Table 1).

Comparison of hemodynamic data and biomarkers in prognostic groups

In the good and poor prognosis groups, there was no significant dif- ference in terms of hemodynamic data, apart from saturation values. The saturation values in the good prognosis group were found to be higher. When lactate medians are compared, patients in the poor prog- nosis group were observed to have higher median values compared to individuals with good prognosis (p b 0.05) (Table 2).

When NT-proBNP values are compared in the good and poor prog- nosis groups, there was a significant difference within 95% confidence intervals (p = 0.040). The median NT-proBNP values were observed to be higher in the poor prognosis group compared to the good progno- sis group. NT-proBNP analyses were only performed for 20 patients due to deficient data (Table 3).

Relationship of prognostic groups to neurological score systems

When the GCS median and mean FOUR scores are compared for pa- tients with good and poor neurologic outcomes, there was a significant difference within 95% confidence interval limits (p b 0.05).

When the variation in FOUR score and GCS over time is investigated, the FOUR score values were significantly different between the progno- sis groups at all times, apart from the 42nd hour (p N 0.05), while the GCS values were not significantly different apart from on admission, and the 3rd and 4th days.

Comparison of prognostic groups with regional cerebral oxygen saturation, CPR duration, body temperature and NT-proBNP

There was no statistically significant difference identified between the prognosis groups in terms of rSO₂, CPR durations, hemoglobin values and admission body temperature (p N 0.05) (Table 3).

When the variation in rSO₂ values over time is investigated, though there was no significant difference between the good and poor progno- sis groups, it appeared to fall in the first 6 h in both prognosis groups (Table 4) (Fig. 1).

When the correlation between rSO₂ values on admission and on the 5th day with NT-proBNP and CPR durations are investigated, there was no significant correlation found (p N 0.05) (Table 5).

Table 2

Descriptive statistics of variables and comparisons by CPC result.

Variables	CPC 1 2				CPC 3 4 5			p
	N	Mean +- sd	Med (Min, Max)		N	Mean +- sd	Med (Min, Max)
Heart rate	7	97.6 +- 10.8	95.9 (86, 119)		18	93,4 +- 14.7	96.2 (51, 117)	0.526t
rSO2	7	69.6 +- 11.3	76.2 (52, 79.5)		18	64.4 +- 11.3	66.7 (41.7,80)	0,312t
Mean arterial pressure	7	86.7 +- 6.6	88 (75.4, 95.6)		18	81.7 +- 8.6	80 (62, 95)	0,18t
Saturation	7	98.9 +- 0.93	99 (97, 99.5)		18	97,1 +- 1.9	97.6 (93, 99)	0,008MW?
Lactate	7	1.5 +- 0,96	0,96 (0.95, 3.6)		18	4.4 +- 6.5	1,96 (1, 27.8)	0,014t?
PaO2	7	138.7 +- 41	140 (88, 192)		18	119.2 +- 29.7	118 (72.8176)	0,199t
PaCO2	7	41.7 +- 3.13	41,4 (35.8, 45.7)		18	43.4 +- 13.06	43,4 (27, 88.5)	0,745MW
Hemoglobin	7	12.2 +- 1,39	12.2 (10.5, 14.1)		18	12.7 +- 2.25	13.1 (7.7,16.3)	0,581t

MW: Mann Whitney U test. t: t-test.

Note: Values in bold are statistically significant.

* p b 0,05.

Table 4

Comparison of the variation of rSO₂ values over time with neurological outcomes.

Variables	Good prognosis		Poor prognosis			Significance
	CPC 1-2 N X+-Sd		CPC 3-5 N	X+-Sd		p	p*
rSO2 admission	7 71.4 +- 12.9		18	69.9 +- 12		0,536t	Within subjects p = 0,227
rSO2 6. hour	7 66.3 +- 8.4		18	66.2 +- 11		0,787t
rSO2 12. hour	7 66.3 +- 11.6		18	66.3 +- 12		0,864t	Between subjects p = 0,491
rSO2 24. hour	7 70.1 +- 15.4		18	66.4 +- 11.9		0,374t
rSO2 30. hour	7 70.3 +- 12.8		18	67.8 +- 10.8		0,374t
rSO2 3.day	7 71.1 +- 12.4		17	65.4 +- 10.7		0,250t
rSO2 4.day	7 71.7 +- 11.9		17	62.1 +- 10.7		0,062t
rSO2 5.day	7 66.9 +- 12.1		16	64.2 +- 10.2		0,626t

t: Independent sample t-test p*: Repeated Measures ANOVA.

There was no significant differences between and within subjects (Time and CPC) p N 0.05).

1. Discussion

In this study researching the efficacy of rSO₂ values in predicting neurologic prognosis of patients surviving after CA in a coma and under- going TTM, rSO₂ was not sufficient to predict good and poor prognosis. Similar to our study, Bougle et al. [1] found no difference between the mean rSO2 values in the good and poor prognosis groups in OHCA pa- tients administered Therapeutic hypothermia . In this study rSO₂ monitoring lasted 48 h, while in our study it lasted 5 days with no sig- nificant difference in the variations over time between the prognosis groups. Similarly, Lemiale et al. found no difference between survivor and non-survivors among patients administered hypothermia on ad- mission to hospital according to cerebral oxygen extraction at 12 and 24 h [11]. In parallel with our study, Meex et al. determined cerebral tis- sue oxygen saturation (SctO₂) in post CA survivor and non-survivor pa- tients were similar at the onset of TH (target temperature 33 ?C/24 h); though contrary to our study after the 3rd hour of TH rates dropped se- verely in Non-survivor patients [9]. In our study though the variation in cerebral oximetry over time was not significantly different in the good and poor neurologic outcome patients, rSO₂ dropped in both groups in the 6th hour of TTM. It has been shown previously in some studies [9,12,13] that SctO₂ reduces with TH. These studies were determined to have a lower target temperature. In our study though cooling was only performed to 35 ?C, there was a reduction observed in the first 6 h. After this fall in the 6th hour, the rSO₂ values in both groups

Fig. 1. Comparison of rSO2 over time with CPC scale.

increased, though this increase was more defined in the good prognosis group. Especially after the 48th hour, while recording a gradual reduc- tion in rSO₂ values in poor neurologic outcome (CPC 3-5) patients, there was a gradual increase recorded in rSO₂ values in good neurologic outcome (CPC 1-2) patients. This attracts attention as seen in the fig- ures, but no statistical significance was identified.

Ito et al. [2] stated that rSO₂ on admission to hospital predicted good neurologic outcome 90 days after OHCA, with patients with good neuro- logic outcomes having higher rSO₂ values. Interestingly in this study 3 patients with good neurologic outcome had very low rSO₂ (<=15%) values. While our study included the variation in rSO₂ over 5 days, Ito et al. performed rSO₂ monitoring for a limited time after admission to hospital. We believe the variation in rSO₂ over time is more important for prognostication of neurologic outcomes.

In our study we determined neurologic scoring systems like the GCS and FOUR score as predictive scales beneficial for prognostication. Though statistically significant for both scales, when the variation in FOUR score over time is investigated we observed it had better prognos- tic Predictive power. The better inter-rater reliability and positive pre- dictive value of FOUR score compared to GCS determined in the literature was correlated in our study [14-16].

BNP is released from the cardiac ventricle in response to increased wall tension. The forerunner of BNP of NT-proBNP provides better prog- nostic information than BNP for patients experiencing myocardial infarctus [17]. NT-proBNP was found to be a good predictive biomarker for determination of neurologic outcomes in our study. The NT-proBNP values in patients with good neurologic outcomes were significantly low. The limited number of articles in the literature on this topic corre- lates with the results of our study. For NT-proBNP, Rhee et al. [18] found it was higher in non-survivor patients, Naveen et al. [19] found it was a useful biological marker for estimation of in-hospital mortality and Mukaida et al. [20] reported it played a major role in estimating outcomes.

Epileptic activity may cause a sympathetic response like hyperten- sion, tachycardia and tachypnea. Claasen et al. [21] reported that during intracortical seizures, along with the sympathetic response, there was an increase in intracranial and cerebral perfusion pressure; however they did not show that global hypermetabolism or brain tissue hypoxia developed. In our study, 7 patients had seizures, with rSO₂ values unaf- fected during epileptic activity as in the study by Meex et al. [9].

Table 5

The correlation between rSO2 values with NT-proBNP and CPR durations.

	NT-proBNP q	p	CPR duration q	p
rSo2 admission	-0,208	0,380	-0,121	0,565
rSo2 5 day	-0,231	0,356	0,038	0,863

Previous studies [22,23] have stated that malignant EEG pattern may have a Predictive role in differentiating good and poor neurologic prog- nosis and be associated with mortality; though they stated that the be- nign pattern did not predict prognosis. In our study, there was no significant correlation between experiencing epileptic seizures and mortality. The reason for this may be that continuous EEG monitoring was not performed. Non-convulsive seizures are not rare in post CA pa- tients [24]. In our study, the lack of EEG assessment may have led to non-convulsive seizures not being recognized.

One of the prognostic factors is duration of CPR. Previous studies have stated that lengthened CPR duration is associated with poor neurologic outcomes. Reynolds et al. [25] showed that the first 10-15 min of conven- tional CPR strategies are most effective and N75% of patients with good functional improvement reach ROSC. Another study [26] revealed that 31.7% of patients with CPR duration b21 min had good neurologic out- come, while in situations N21 min extracorporeal CPR should be consid- ered. In our study, there was no significant difference between good and poor neurologic outcomes in terms of CPR durations. Our CPR dura- tions were mainly in the interval from 10 to 20 min. The reason for the lack of difference may be due to the small sample and the correlation with above mentioned studies with most of our CPR durations b20 min. Studies investigating the correlation between hospital admission body temperature (ABT) in post CA patients and mortality stated that hypothermia was an indicator of poor neurologic outcomes [27,28]. One of these studies took the hypothermia cut-off as 34 ?C, while the other study used 35 ?C. There was no significant difference found for ABT values above 35 ?C. In our study ABT had no effect on neurologic outcome. In terms of ABT in our data there was no value below 35 ?C

which shows our findings comply with the literature.

1. Limitations

Ours study has several limitations. First, we used a single-center de- sign and a small patient population. Second, together with variations in rSO₂, continuous monitoring of cerebral hemodynamic parameters (with transcranial Doppler), SjvO₂ and brain tissue oxygen tension mon-

itoring were not performed. Though associated with SjvO , rSO was mea-

1. Ito N, Nishiyama K, Callaway CW, Orita T, Hayashida K, Arimoto H, et al. Noninvasive regional cerebral oxygen saturation for neurological prognostication of patients with out-of-hospital cardiac arrest: a prospective Multicenter observational study. Resus- citation 2014 Jun;85(6):778-84. https://doi.org/10.1016/j.resuscitation.2014.02.012 (Epub 2014 Mar 5).
2. Kirkegaard H, Soreide E, de Haas I, Pettila V, Taccone FS, Arus U, et al. Targeted tem- perature management for 48 vs 24 hours and neurologic outcome after out-of-hos- pital cardiac arrest. JAMA 2017 Jul 25;318(4):341-50. https://doi.org/10.1001/jama. 2017.8978.
3. Nishiyama K, Ito N, Orita T, Hayashida K, Arimoto H, Beppu S, et al. Regional cerebral oxygen saturation monitoring for predicting interventional outcomes in patients fol- lowing out-of-hospital cardiac arrest of presumed cardiac cause: a prospective, ob- servational, multicentre study. Resuscitation 2015 Nov;96:135-41. https://doi.org/ 10.1016/j.resuscitation.2015.07.049 (Epub 2015 Aug 17).
4. Neumar RW, Nolan JP, Adrie C, Aibiki M, Berg RA, Bottiger BW, et al. Post-cardiac ar- rest syndrome: epidemiology, pathophysiology, treatment, and prognostication. A consensus statement From the International Liaison Committee on Resuscitation (American Heart Association, Australian and New Zealand Council on Resuscitation, European Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Asia, and the Resuscita- tion Council of Southern Africa); 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; and the Stroke Council. Circulation 2008 Dec 2;118 (23):2452-83. https://doi.org/10.1161/CIRCULATIONAHA.108.190652 (Epub 2008

Oct 23).

1. Callaway CW, Donnino MW, Fink EL, et al. Part 8: post-cardiac arrest care: 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2015;132(18):S465-82 (suppl 2).
2. Nolan JP, Soar J, Cariou A, Cronberg T, Moulaert VR, Deakin CD, et al. European resus- citation council and European society of intensive care medicine 2015 guidelines for post-resuscitation care. Intensive Care Med 2015 Dec;41(12):2039-56.
3. Group HACAS. Mild therapeutic hypothermia to improve the neurological outcome after cardiac arrest. N Engl J Med 2002;346:549-65.
4. Meex I, Dens J, Jans F, Boer W, Vanhengel K, Vundelinckx G, et al. Cerebral tissue ox- ygen saturation during therapeutic hypothermia in post-cardiac arrest patients. Re- suscitation 2013 Jun;84(6):788-93. https://doi.org/10.1016/j.resuscitation.2013.01. 003 (Epub 2013 Jan 8).
5. Ikeda Keita, MacLeod David B, Grocott Hilary P, Moretti Eugene W, Ames Warwick, Vacchiano Charles. The accuracy of a Near-infrared spectroscopy cere- bral oximetry device and its potential value for estimating jugular venous oxy- gen saturation. Anesth Analg 2014 Dec;119(6):1381-92. https://doi.org/10. 1213/ANE.0000000000000463.
6. Lemiale V, Huet O, Vigue B, et al. Changes in cerebral blood flow and oxygen extrac- tion during post-resuscitation syndrome. Resuscitation 2008;76:17-24.
7. Tobias JD, Russo P, Russo J. Changes in near infrared spectroscopy during deep hypo- thermic circulatory arrest. Ann Card Anaesth 2009;12:17-21.
8. Joshi B, Brady K, Lee J, et al. Impaired autoregulation of cerebral blood flow during

2 2 rewarming from hypothermic cardiopulmonary bypass and its potential association

sured from the forehead region using NIRS technology. This only provides information about the frontal region. Multimodal neuromonitoring would have allowed continuous assessment of rSO₂ in addition to better comprehension of cerebral hemodynamic disorders. Third, we did not compare the increase in neuron specific enolase with the cerebral oxime- try prognostic value. Fourth, the initial time of beginning TTM and pre- induction delays were not assessed in our study.

1. Conclusion

In our study we did not identify a significant correlation between rSO₂ values and neurologic outcomes. We could not obtain a clear idea of rSO₂ prediction of neurologic prognosis, but variations over time and the variation rate in the first 6 h may provide prediction in terms of neurologic prognostication. Additionally, NT-proBNP may be useful in the prediction of neurologic prognosis after CA. We believe it would be more accurate to use multimodal monitoring methods like the vari- ation in rSO₂, along with pro BNP, FOUR score and lactate levels.

Conflicting interest

No conflict of interest was declared by the authors.

References

[1] Bougle A, Daviaud F, Bougouin W, Rodrigues A, Geri G, Morichau-Beauchant T, et al. Determinants and significance of cerebral oximetry after cardiac arrest: a prospec- tive cohort study. Resuscitation 2016 Feb;99:1-6. https://doi.org/10.1016/j.resusci- tation.2015.11.011 (Epub 2015 Nov 30).

with stroke. Anesth Analg 2010;110:321.

1. Khajeh Ali, Fayyazi Afshin, Miri-Aliabad Ghasem, Askari Hasan, Noori Noormohammad, Khajeh Behrouz. Comparison between the ability of Glasgow coma scale and full outline of Unresponsiveness score to predict the mortality and discharge rate of pediatric intensive care unit patients. Iran J Pediatr 2014 Oct;24

(5):603-8 (Published online 2014 Sep 12).

1. Fischer Michael, Ruegg Stephan, Czaplinski Adam, Strohmeier Monika, Lehmann Angelika, Tschan Franziska, et al. Inter-rater reliability of the full outline of unre- sponsiveness score and the Glasgow coma scale in critically ill patients: a prospec- tive observational study. Crit Care 2010;14(2):R64. https://doi.org/10.1186/cc8963 (Published online 2010 Apr 14).
2. Khanal Kishor, Bhandari Sanjeeb Sudarshan, Shrestha Ninadini, Acharya Subhash Prasad, Marhatta Moda Nath. Comparison of outcome predictions by the Glasgow coma scale and the full outline of unresponsiveness score in the neurological and neurosurgical patients in the intensive care unit. Indian J Crit Care Med 2016 Aug; 20(8):473-6. https://doi.org/10.4103/0972-5229.188199.
3. Maisel AS, Krishnaswamy P, Nowak RM, McCord J, Hollander JE, Duc P, et al. Rapid measurement of B-type natriuretic peptide in the emergency diagnosis of heart fail- ure. N Engl J Med 2002;347:161-7. https://doi.org/10.1056/NEJMoa020233.
4. Rhee Chin Kook, So Yeon Lim, Koh Shin Ok, Choi Won-Il, Lee Young-Joo, Chon Gyu Rak. Usefulness of N-terminal pro-B-type natriuretic peptide in patients admitted to the intensive care unit: a Multicenter prospective observational study. BMC Anesthesiol 2014;14:16. https://doi.org/10.1186/1471-2253-14-16.
5. Naveen Vandanapu, Vengamma Bhuma, Mohan Alladi, Vanajakshamma Velam. N- terminal pro-brain natriuretic peptide levels and short term prognosis in acute is- chemic stroke. Ann Indian Acad Neurol 2015 Oct-Dec;18(4):435-40. https://doi. org/10.4103/0972-2327.165478.
6. Mukaida Hiroshi, Hayashida Masakazu, Matsushita Satoshi, Yamamoto Makiko, Nakamura Atsushi, Amano Atsushi. Brain natriuretic peptide may play a major role in risk stratification based on cerebral oxygen saturation by near- infrared spectroscopy in patients undergoing major cardiovascular surgery. PLoS One 2017;12(7):e0181154. https://doi.org/10.1371/journal.pone.0181154 (Pub- lished online 2017 Jul 12).
7. Claassen Jan, Perotte Adler, Albers David, Kleinberg Samantha, Michael Schmidt J, Bin Tu, et al. nonconvulsive seizures after subarachnoid hemorrhage: multimodal detection and outcomes. Ann Neurol 2013 Jul;74(1):53-64. https://doi.org/10. 1002/ana.23859.
8. Fugate JE, Wijdicks EF, Mandrekar J, Claassen DO, Manno EM, White RD, et al. Predic- tors of neurologic outcome in hypothermia after cardiac arrest. Ann Neurol 2010;68: 907-14.
9. Taccone Fabio Silvio, Cronberg Tobias, Friberg Hans, Greer David, Horn Janneke, Oddo Mauro, et al. How to assess prognosis after cardiac arrest and therapeutic hy- pothermia. Crit Care 2014;18(1):202. https://doi.org/10.1186/cc13696 (Published online 2014 Jan 14).
10. Rittenberger Jon C, Popescu Alexandra, Brenner Richard P, Guyette Francis X, Callaway Clifton W. Frequency and timing of nonconvulsive status epilepticus in co- matose post-cardiac arrest subjects treated with hypothermia. Neurocrit Care 2012; 16(1):114-22. https://doi.org/10.1007/s12028-011-9565-0.
11. Reynolds JC, Frisch A, Rittenberger JC, Callaway CW. Duration of resuscitation efforts and functional outcome after out-of-hospital cardiac arrest: when should we change to novel therapies? Circulation 2013 Dec 3;128(23):2488-94. https://doi.org/10. 1161/CIRCULATIONAHA.113.002408 (Epub 2013 Nov 17).
12. Jin Kim Su, Jung Jae Seung, Park Jae Hyoung, Park Jong Su, Hong Yun Sik, Lee Sung Woo. An optimal transition time to extracorporeal cardiopulmonary resuscitation for predicting good neurological outcome in patients with out-of-hospital cardiac ar- rest: a propensity-matched study. Crit Care 2014;18(5):535. https://doi.org/10. 1186/s13054-014-0535-8 (Published online 2014 Sep 26).
13. Hovdenes J, Roysland K, Nielsen N, Kjaergaard J, Wanscher M, Hassager C. A low body temperature on arrival at hospital following out-of-hospital-cardiac-arrest is associated with increased mortality in the TTM-study. Resuscitation 2016;107: 102-6.
14. den Hartog Alexander W, de Pont Anne-Cornelie JM, Robillard Laure BM, Binnekade Jan M, Schultz Marcus J, Horn Janneke. Spontaneous hypothermia on intensive care unit admission is a predictor of unfavorable neurological outcome in patients after resuscitation: an observational cohort study. Crit Care 2010;14(3):R121. https:// doi.org/10.1186/cc9077 (Published online 2010 Jun 23).