a b s t r a c t

Objective: The objective of the study is to demonstrate the utility of Near-infrared spectroscopy (NIRS) in evaluating chest compression quality in cardiac arrest patients as well as determine its prognosis predictive value.

Methods: We present a nonconsecutive case series of adult patients with CA whose cardiopulmonary resuscitation (CPR) was monitored with NIRS and collected the total hemoglobin concentration change (?cHb), the tissue oxygen index (TOI), and the ?TOI to assess CC quality in a noninvasive fashion.

Results: During CPR, ?cHb displayed waveforms monitor, which we regarded as a surrogate for CC quality. Total hemoglobin concentration change waveforms responded accurately to variations or cessations of CCs. In addition, a TOI greater than 40% measured upon admission appears to be significant in predicting patient’s outcome. Of 15 patients, 6 had a TOI greater than 40% measured upon admission, and 67% of the latter were in return of spontaneous circulation after CPR and were found to be significantly different between return of spontaneous circulation and death (P = .047; P b .05).

Conclusion: Near-infrared spectroscopy reliably assesses the quality of CCs in patients with CA demonstrated by synchronous waveforms during CPR and possible prognostic predictive value, although further investigation is warranted.

(C) 2013

Introduction

Cardiac arrest decreases brain Tissue oxygenation, which eventually causes neurologic damage [1]. Cardiopulmonary resusci- tation (CPR) is the standard procedure to CA, although survival outcomes still remain astonishingly poor [2].

Despite frequent updates from international CPR guidelines, it is still common that even highly trained hospital personnel perform inadequate Chest compressions , which consequently decrease cerebral perfusion pressure [3]. Therefore, the cardinal requisite of quality CPR, namely, effective CC, deserves particular scrutiny to improve manual Resuscitation quality and performance.

Although several instrumentation methods are available to quantify coronary artery blood flow: end-tidal CO₂, “smart de- fibrillators” that provide feedback on chest compression rate and depth as well as diastolic Blood pressure measurements [3], there is yet to be a handy “gold standard” noninvasive method that evaluates cerebral blood volume and flow. Near-infrared spectroscopy (NIRS) may be the appropriate candidate to resolve this issue (Fig. 1).

* Corresponding author. Tel.: +81 44 977 8111.

E-mail address: [email protected] (Y. Koyama).

Near-infrared spectroscopy has useful applications in the clinical scenery: cardiopulmonary bypass, neurosurgery, pediatric intensive care, carotid endarterectomy, and trauma [4,5].

However, no study has yet determined a positive correlation between quality-controlled CPR and tissue oxygen index (TOI) as well as total hemoglobin concentration change (?cHb) (curve ampli- tudes), synchronously with compressions.

We set out to evaluate CPR quality by comparing ?cHb readings with each CC performed and assessed whether sufficient changes of amplitude of ?cHb indicated CPR quality, hence presumably demon- strating chest compression effectiveness.

We also investigated whether TOI readings reflecting increased cerebral perfusion with each effective compression had any predictive value evidenced by return of spontaneous circulation (ROSC).

Methods

The prospective, nonconsecutive observational study was ap- proved by the ethics committee of St Marianna University School of Medicine, a 1200-bed teaching hospital in Kawasaki, Japan. We obtained informed written consent from the next of kin of 15 CA patients (10 males and 5 females) who presented to our emergency department (ED). The patient population mean age was 79 years

0735-6757/$ - see front matter (C) 2013 http://dx.doi.org/10.1016/j.ajem.2013.07.002

near-infrared spectroscopy device“>Fig. 1. NIRO (Hamamatsu Photonics, Japan) uses NIRS. The device consists of a screen for real-time monitoring, 2 bilateral frontal probes.

(range, 55-99 years). Exclusion criteria were younger than 18 years and trauma patients.

Near-infrared spectroscopy device

NIRO (Hamamatsu Photonics, Hamamatsu-shi Shizuoka, Japan) uses an NIRS, a noninvasive, real-time continuous monitoring algorithm that reliably estimates oxygen delivery through biological tissues. Following the principle of optical spectroscopy, within the range of 700 to 1300 nm, NIRS probes containing a laser diode irradiate narrow transillumination wavelengths several centimeters deep into tissue and are primarily absorbed by metalloprotein complex chromophores, namely, hemoglobin as both oxyhemoglobin and deoxyhemoglobin (HbO₂ and HHb, respectively) [4,6].

Although subject for a separate report, NIRS assesses the concentration changes in oxygenated (?HbO₂), deoxygenated (?HHb), and total tissue hemoglobin (?cHb = ?HbO2 + ?HHb) by the Modified Beer-Lambert method with units expressed in micro- mole per liter. Total hemoglobin concentration change represents the change of cerebral blood volume within the segment of tissue analyzed and its waveform displayed on monitor and evaluates cerebral perfusion and thus CC [5,7].

On the other hand, NIRS also measures Tissue hemoglobin oxygen saturation in the cerebral microvasculature (TOI - tissue oxygenated index), which is the ratio of oxygenated to total hemoglobin (TOI = HbO2/cHb) expressed in absolute percentage value (%), using the spatially resolved spectroscopy method.

The device contains 2 probes that are applied bilaterally on the patients’ supraorbital region 2 cm above the eyebrows (Fig. 2A). Each probe contains an near-infrared (NIR) transillumination source (emitter), which irradiates through the skin, and a separate photodiode (detector) that detects light returning from tissue and processes data to be continuously displayed in real time on monitor. The optimal emitter/detector distance of 4 cm increases the depth of NIR light penetration that will follow an elliptical trajectory approximately 2 cm deep within cerebral tissue to assess cerebral oxygenation [8] (Fig. 2B).

Study protocol

Within arrival to ED, emergency medical technicians relay CPR to emergency clinicians, who will sustain CPR until ROSC or death.

Fig. 2. A and B, The NIRS device probes are applied bilaterally on the patient’s forehead. The NIR-transmitted light traversing through the skull will penetrate through various layers of tissue and will follow an elliptical pathway.

At that time, sensors are applied to patient’s supraorbital region to measure TOI and ?cHb. Tissue oxygen index is displayed continuously but recorded twice: (1) within 3 minutes or less upon admission to ED (TOIadm) and (2) upon time of ROSC acquisition or time of death when CPR is discontinued (TOIcpr), whereas ?TOI is rapidly tabulated from the gathered data (?TOI = TOIcpr - TOIadm). Mean arterial pressure, end-tidal CO₂, electrocardiogram, systolic blood pressure, and pulse oximetry were continuously evaluated on separate device.

Calibration

A time sampling calibration of 2-second represents the average of all CCs performed every 2 seconds but only detects ?cHb in 2-second interval, which does not represent the response of every CC. Therefore, recalibrating the sampling time from 2 to 0.05 seconds amplified the signals by shortening the time interval between measurements (40-fold). A 0.05-second interval sampling assessed CPR with more conformity and fidelity. Waveform amplitudes could then be displayed with greater accuracy and reflected with more exactness the actual response of continuous CPR in real time, which would not be detectable otherwise.

Near-infrared spectroscopy displayed accurate cerebral blood volume waveform in synchrony with CC after recalibrating the time sample from 2 to 0.05 seconds.

In this study, we used statistical analysis using a 2 x 2 contingency table (Fisher direct probability test) to determine whether any correlation between ?TOI, TOIadm, and TOIcpr with ROSC existed within the 15-patient population. Although effective CC could not be quantifiably measured and, consequently, no direct correlation to ROSC could be established, we, nevertheless, sought to determine

cerebral hemodynamics“>whether effective CC, upon visual inspection, correlated with

?cHb waveforms.

Results

Table describes the demographic and data gathered for all the 15 patients admitted to our ED with CA, whom we assessed ?cHb and TOI with NIRS, simultaneous to Manual CPR over a 1-year period (October 2010 to February 2012).

Of the 15 CA patients, 7 (47%) had a ?TOI 10% or greater, and of these, 4 (57%) were ROSC. Notably, only 1 patient (12.5%) with ?TOI less than 10% was ROSC (Fig. 3A).

Similarly, 6 (40%) of the 15 patients had a TOIadm 40% or greater and TOIcpr 50% or greater, of whom 4 (67%) were ROSC, whereas 1 patient (11%) with TOIadm less than 40% and TOIcpr less than 50% was ROSC (Fig. 3B and C).

Although no significant difference was determined between ?TOI 10% or greater and ROSC (P = .100; P N .05), we determined significant differences between both TOIadm 40% or greater and ROSC (P = .047; P b .05) as well as TOIcpr 50% or greater and ROSC (P = .047; P b .05).

Discussion

Near-infrared spectroscopy monitors cerebral oxygenation, pro- vides accurate quality assessment of chest compression, and predicts the ROSC of CA patients during resuscitation.

Cerebral hemodynamics

The positive relation between cardiac output and cerebral blood flow (CBF) [9] implicates that manual CPR also affects CBF and may be quantified by NIRS. For each compression, NIRS evaluates the ?cHb and interprets the data into CC waveforms. Their varying shapes and amplitudes represent CBF fluctuation that was only visually identified from the monitor screen. We assumed these waveforms to correlate with ?cHb and hence serve to assess CC quality, as occurred in patient

1. The waveforms ceased when CC was discontinued and progres- sively resumed once reinitiated (Fig. 4A). This information is significant in adjusting depth of compression applied.

Fig. 3. A, B, and C, Patients status according ?TOI, TOIadm, and TOIcpr.

Cerebral perfusion pressure (arterial pressure - intracranial pressure), cerebral venous oxygen saturation monitoring, transcra- nial Doppler, and positron emission tomography are also alternate methods to assess cerebral blood volume/flow but all appear to contain impracticalities when dealing with the emergent patient. Indeed, the arterial pressure, intracranial pressure, and cerebral venous oxygen saturation monitoring are parameters that can only be obtained invasively. In addition, continuously holding the transcranial Doppler probe during CC can be difficult to manage, and finally, acquiring images from the time-consuming positron emission tomography scan may put the unstable patient at risk [10,11].

Table

Patients demographic, vitals, and data.

Patient	Age	Sex	CPR	CPR	ECG monitor	BP after	TOI (%)				?cHb (umol/L)		ROSC
			EMT	ED to		ROSC							or
			to ED	ROSC	On arrival		TOIadm	TOIcpr	?TOI	Upon arrival CPR is discontinued			death
			arrival	or									in ED
			(min)	death
				(min)
1	66	M	22	30	VF	63/47	47	42	-5	0.6		0.82	ROSC
2	79	M	40	66	Asystole	-	37	38	1	0.46		0.34	Deatha
3	55	M	44	76	PEA	-	31	44	13	0.38		0.61	Deatha
4	73	M	39	55	Asystole	-	23	27	4	0.9		0.51	Deatha
5	99	F	50	69	PEA	-	30	33	3	0.5		0.42	Deatha
6	81	M	5	25	PEA	81/57	45	83	38	0.66		0.2	ROSC
7	88	M	47	61	Asystole	83/56	49	64	15	0.77		0.15	ROSC
8	67	F	40	60	Asystole	53/37	38	52	14	0.59		0.2	ROSC
9	86	F	28	50	Asystole	-	28	21	-7	0.8		0.45	Deatha
10	85	M	4	15	PEA	176/123	39	45	6	1.5		0.48	Deatha
11	85	M	37	52	Asystole	48/23	46	57	11	0.51		1.57	ROSC
12	84	F	20	37	PEA	-	39	40	1	0.73		0.55	Deatha
13	77	M	46	63	PEA	-	28	34	6	1.45		0.82	Deatha
14	82	F	42	52	PEA	-	41	56	15	0.62		0.38	Deatha
15	88	M	39	46	Asystole	125/81	51	70	19	0.73		0.38	Deatha

Abbreviations: M, male; F, female; VF, ventricular fibrillation; PEA, pulseless electrical activity.

^a Patient died in the ED during CPR.

Fig. 4. A, Chest compressions of varying intensity and depth are depicted as ?cHb waveforms with varying amplitudes. Interruption of CC causes waveforms to disappear and reappear when CCs are reinstated. B, Chest compressions greater than 5 cm provide sufficient cerebral oxygenation, whereas compressions less than 5 cm do not provide sufficient cerebral oxygenation. C, Same patient after ROSC.

Unlike peripheral pulse oximetry, NIRS continues to work in the absence of pulsatile blood flow and can continuously measure cerebral oxygenation in hypotensive or patients in circulatory arrest [12].

Cerebral tissue oxygenation and Prognostic prediction

Near-infrared spectroscopy monitors cerebral tissue oxygenation (TOI) with high sensitivity and specificity [13], which is computed from the venous (70%) blood compartment. Eighty-five percent of cerebral oxygen saturation is derived from cortical tissues, which justifies probes placements on the forehead [4]. Presumably, these relationships persist in the setting of CA.

Ito et al [14] measured the regional cerebral oxygenation saturation (rSO₂) with a different NIRS device (INVOS, Somanetics, MI) on CA patients who had arrived within 3 minutes to the ED. They reported any rSO₂ of 15% or less to be indicative of a Poor neurologic outcome, whereas rSO₂ of 25% or greater could predict a positive outcome.

Although, no definitive threshold TOI defining positive outcome has yet been agreed upon, we suspected that a TOIadm 40% or greater, TOIcpr 50% or greater or a ?TOI 10% or greater might predict a positive prognosis as occurred in patient 7. The subject, an 88- year-old man with asystole, indicated a TOIadm of 49%. The asystole was contained after 61 minutes of CPR, the subject regained ROSC, and we noted increased ?cHb waveforms on monitor. The TOIcpr increased up to 64% from the original 49%, which represents a ?TOI 10% or greater (?TOI, 15%). The patient’s BP was maintained at 83/

56 mm Hg and remained conscious for the next 24 hours but passed away from heart failure 1 day later (Fig. 4C). We presumed that the subject’s elevated TOIadm 40% or greater may be, in part, the result of continuous CPR engaged by Emergency Medical Service from the moment of contact until delivery to ED. Out-of-hospital CPR significantly increases survival rate, and to this effect, patient outcome may largely depend on its efficiency [15]. Near-infrared spectroscopy appears to fit Emergency Medical Service demands as it improves CPR efficiency, and by maintaining an elevated TOIadm 40% or greater during prehospital CPR, it could play an essential role in successful resuscitation. A TOIadm 40% or greater could then predict positive prognosis, although further investigation is warranted.

The purpose of our study limited itself on CPR quality assessment, and we therefore considered CPR to be “effective” in patients who had attained ROSC and “ineffective” in those who did not reach ROSC, regardless of patient’s final outcome.

Some authors identified no significant correlation between high- quality CPR and cerebral oxygenation with the use of NIRS (INVOS 5100c) [16], whereas others have demonstrated on animal models that measuring regional cerebral tissue oxygenation with NIRS could assess CPR effectiveness [17]. Our study demonstrated that NIRS displayed waveforms as well as TOI variations in synchrony with each

CC. Compared with TOIadm, TOIcpr increased in 13 patients (87%), confirming increased cerebral oxygenation and, therefore, effective CPR. However, a ?TOI 10% or greater appeared not to hold prognostic value as originally expected, although TOIadm 40% or greater and TOIcpr 50% or greater demonstrated to be significantly correlated with ROSC and, hence, positive predictive value.

Limitations

There are several limitations of this study, including the small number of patients and the lack of HbO2 and HHb data. Obtaining such data would provide additional value to NIRS’s usefulness in evaluating CPR quality.

Despite the positive correlation between high-quality CPR, CBF, and tissue oxygenation observed in this study, it is necessary to address that all our patients died, which is a significant limitation and may reflect procedural mismanagement, although our sole aim was to assess CPR quality.

Future application

The utility of NIRS deserves strong consideration in the out-of- hospital CPR ambulatory setting because TOIadm 40% or greater has positive predictive value. It can play an important role in monitoring CPR quality, providing reliable feedback, and assisting the EMT address the appropriate course of management to pursue. Near- infrared spectroscopy in such new context can increase survival rates of out-of-hospital CAs, although further investigation regarding its practicality should be undertaken.

Conclusion

Our data suggest that during quality-controlled CPR, a TOIcpr 50% or greater and/or a TOIadm of 40% or greater correlate significantly with ROSC and may predict positive prognosis. This preliminary study identifies the importance of quality prehospital CPR with NIRS, which provides cerebral perfusion and tissue oxygenation feedback infor- mation as a surrogate for the Quality of CPR administered. Near- infrared spectroscopy (NIRO) appears as a valuable method when administering CPR on CA patients whether in the ward, intensive care unit, or outside the hospital.

Acknowledgment

The authors wish to thank Shinobu Tatsunami PhD for statistical instructions in this manuscript.

References

1. Safar P. Cerebral resuscitation after cardiac arrest: a review. Circulation 1986;74: 138-53.
2. Peberdy MA, Kaye W, Ornato JP, et al. Cardiopulmonary resuscitation of adults in the hospital: a report of 14720 cardiac arrests from the National Registry of Cardiopulmonary Resuscitation. Resuscitation 2003;58:297-308.
3. Abella BS, Sandbo N, Vassilatos P, et al. Chest compression rates during cardiopulmonary resuscitation are suboptimal: a prospective study during in- hospital cardiac arrest. Circulation 2005;111(4):428-34.
4. Murkin JM, Arango M. Near-infrared spectroscopy as an index of brain and tissue oxygenation. Br J Anaesth 2009;103:3-13.
5. Ward KR, Ivatury RR, Barbee RW, et al. Near infrared spectroscopy for evaluation of the trauma patient: a technology review. Resuscitation 2006;68(1):27-44.
6. Jobsis FF. Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 1977;198:1264-7.
7. Nagdyman N, Fleck TP, Ewert P, et al. Cerebral oxygenation measured by near- infrared spectroscopy during circulatory arrest and cardiopulmonary resuscita- tion. Br J Anaesth 2003;91(3):438-42.
8. Suzuki S, Takasaki S, Ozaki T, et al. Tissue oxygenation monitor using NIR spatially resolved spectroscopy. Optical Tomography and Spectroscopy of Tissue III. Proc SPIE 1999;359(7):582.
9. Kusaka T, Okubo K, Nagano K, et al. Cerebral distribution of cardiac output in newborn infants. Arch Dis Child Fetal Neonatal Ed 2005;90:77-8.
10. Yannopoulos D, McKnite S, Aufderheide TP, et al. Effects of incomplete chest wall decompression during cardiopulmonary resuscitation on coronary and cerebral perfusion pressures in a porcine model of cardiac arrest. Resuscitation 2005;64(3):363-72.
11. Blumenstein J, Kempfert J, Walther T, et al. Cerebral flow patteRN monitoring by transcranial Doppler during cardiopulmonary resuscitation. Anaesth Intensive Care 2010;38(2):376-80.
12. Pilkington SN, Hett DA, Pierce JMT, et al. Auditory evoked responses and near- infrared spectroscopy during cardiac arrest. Br J Anaesth 1995;74:717-9.
13. Al-Rawi PG, Smielewski P, Kirkpatrick PJ. Evaluation of a Near-Infrared Spectrometer (NIRO 300) for detection of intracranial oxygenation changes in the adult head. Stroke 2001;32:2492-500.
14. Ito N, Nanto S, Nagao K, Hatanaka T, Kai T. Regional cerebral oxygen saturation predicts poor neurological outcome in patients with out-of-hospital cardiac arrest. Resuscitation 2010;81(12):1736-7.
15. Valenzuela TD, Roe DJ, Cretin S, et al. Estimating effectiveness of cardiac arrest interventions: a logistic regression survival model. Circulation 1997;96:3308-13.
16. Kamarainen A, Sainio M, Olkkola KT, et al. Quality controlled manual chest compressions and cerebral oxygenation during in-hospital cardiac arrest. Resuscitation 2012;83(1):138-42.
17. Chien JC, Jeng MJ, Chang HL, et al. Cerebral oxygenation during hypoxia and resuscitation by using near-infrared spectroscopy in newborn piglets. J Chin Assoc 2007;70(2):47-55.