Neurologic Critical Care
A Feasibility Study of Cerebral Oximetry During In-Hospital Mechanical and Manual Cardiopulmonary Resuscitation* Sam Parnia, MD, PhD; Asad Nasir, MD; Anna Ahn, MD; Hanan Malik, MD; Jie Yang, PhD; Jiawen Zhu, MS; Francis Dorazi, DO; Paul Richman, MD
Objective: A major hurdle limiting the ability to improve the quality of resuscitation has been the lack of a noninvasive real-time detection system capable of monitoring the quality of cerebral and other organ perfusion, as well as oxygen delivery during cardiopulmonary resuscitation. Here, we report on a novel system of cerebral perfusion targeted resuscitation. Design: An observational study evaluating the role of cerebral oximetry (Equanox; Nonin, Plymouth, MI, and Invos; Covidien, Mansfield, MA) as a real-time marker of cerebral perfusion and oxygen delivery together with the impact of an automated mechanical chest compression system (Life Stat; Michigan Instruments, Grand Rapids, MI) on oxygen delivery and return of spontaneous circulation following in-hospital cardiac arrest. Setting: Tertiary medical center. Patients: In-hospital cardiac arrest patients (n = 34). Main Results: Cerebral oximetry provided real-time information regarding the quality of perfusion and oxygen delivery. The use of automated mechanical chest compression device (n = 12) was associated with higher regional cerebral oxygen saturation compared with manual chest compression device (n = 22) (53.1% ± 23.4% vs 24% ± 25%, p = 0.002). There was a significant difference in mean regional cerebral oxygen saturation (median % ± interquartile range) in patients who achieved return of sponta-
*See also p. 1001. All authors: Resuscitation Research Group, Division of Pulmonary & Critical Care, Department of Medicine, State University of New York at Stony Brook, Stony Brook Medical Center, T17-040 Health Sciences Center, Stony Brook, NY. Supported, in part, by Stony Brook Medical Center, Department of Medicine Pilot Project Program Award. The authors have disclosed that they do not have any potential conflicts of interest. Address requests for reprints to: Sam Parnia, MD, PhD, MRCP, State University of New York at Stony Brook, Stony Brook Medical Center, T17040 Health Sciences Center, Stony Brook, NY 11794–8172. E-mail: sam. [email protected] Copyright © 2013 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0000000000000047
930
www.ccmjournal.org
neous circulation (n = 15) compared with those without return of spontaneous circulation (n = 19) (47.4% ± 21.4% vs 23% ± 18.42%, p < 0.001). After controlling for patients achieving return of spontaneous circulation or not, significantly higher mean regional cerebral oxygen saturation levels during cardiopulmonary resuscitation were observed in patients who were resuscitated using automated mechanical chest compression device (p < 0.001). Conclusions: The integration of cerebral oximetry into cardiac arrest resuscitation provides a novel noninvasive method to determine the quality of cerebral perfusion and oxygen delivery to the brain. The use of automated mechanical chest compression device during in-hospital cardiac arrest may lead to improved oxygen delivery and organ perfusion. (Crit Care Med 2014; 42:930–933) Key Words: cardiac arrest; cerebral oximetry; cerebral perfusion; near-infrared spectroscopy; resuscitation; return of spontaneous circulation
A
lthough in-hospital resuscitation leads to return of spontaneous circulation (ROSC) in up to 50%, survival to discharge remains poor at 18% (1). Pathophysiologically, cardiac arrest reflects an injury process that begins with the cessation of heartbeat and reflects a state of imbalanced oxygen delivery relative to tissue oxygen requirements, which if left unrecognized and untreated can lead to progressive and irreversible organ damage (2). Even though one of the major goals of cardiopulmonary resuscitation (CPR) is maintaining adequate cerebral perfusion and oxygen delivery, to date there have been no standard noninvasive “real-time” clinical methods available to determine the balance between oxygen delivery and requirements in the brain during cardiac arrest resuscitation. Although jugular venous oxygen saturation (Sjo2) monitoring can determine the balance between cerebral oxygen delivery and oxygen requirements, it is an invasive test, which greatly limits its practical application during cardiac arrest resuscitation. A noninvasive technology that has emerged and is not susceptible to motion artifact during cardiac arrest is cerebral oximetry using near-infrared spectroscopy (NIRS). April 2014 • Volume 42 • Number 4
Neurologic Critical Care
NIRS measures the fall in optical intensity, expressed as the change in optical density per centimeter of tissue, and when applied to the brain, it provides real-time continuous information regarding regional cerebral oxygen saturation (rSo2) in the frontal lobe, which reflects the balance between oxygen delivery and consumption in cerebral tissue (3). Although the quality of chest compression delivery is of prime importance during cardiac arrest resuscitation, many studies have demonstrated significant variability in the delivery of chest compressions in hospitals (4–6). Analysis of the quality of CPR using a quantitative system of CPR (Real CPR; Zoll, Chelmsford, MA) during simulation training at our institution demonstrated that staff were able to achieve the rate and depth recommended by the American Heart Association guidelines less than 50% of the time. However, the use of an automated mechanical chest compression (A-CPR) device (Life Stat; Michigan Instruments, MI) demonstrated the correct delivery of depth and rate 100% of the time. Thus, to standardize the delivery of chest compression across the hospital, we introduced A-CPR. Furthermore, building on an earlier feasibility study in which NIRS was incorporated into a small series of in-hospital manual cardiac arrest resuscitations at our institution (7), we compared the quality of cerebral oxygen delivery using A-CPR device with conventional manual chest compression (M-CPR) device at our specific institution. We hypothesized that standardizing the rate and depth of CPR delivery using A-CPR would lead to improved brain oxygen delivery. The primary outcome measure for this study was the mean rSo2 level during CPR in the A-CPR group compared with the M-CPR group.
MATERIALS AND METHODS Institutional review board approval was obtained for this study following modification to the clinical method used to resuscitate cardiac arrest patients at Stony Brook Medical Center. We conducted an observational study in a convenience sample of 34 adult patients (age > 21) who underwent in-hospital cardiac arrest resuscitation during a 1-year clinical evaluation period. This was set up to examine the utility of incorporating cerebral oximetry using NIRS (Equanox 7600; Nonin, Plymouth, MI, and Invos; Somanetics, Troy, MI) during cardiac arrest. Each sensor consists of an adhesive strip, a near-infrared light transmitter, and two sensors. The emitted light is scattered by tissues in two parabolic curves: one corresponding with hemoglobin saturation from the skin and skull and the other from the skin, skull, and frontal cortex. Using specific algorithms, the hemoglobin saturation in the frontal cortex is calculated (8–11). After a cardiac arrest was announced through the paging system in the hospital, the cerebral oximeter was taken by a predesignated member of staff from the medical ICU (MICU) to the site of cardiac arrest, where an on-call MICU physician and nurse trained in its use also attended. The placement of a single sensor on either the right or left lateral aspect of the forehead was considered sufficient for the purpose of measuring the overall quality of cerebral oxygen delivery, since Critical Care Medicine
cerebral blood flow during cardiac arrest is predominantly dependent on the quality of the circulation. The cerebral oximeter recorded continuous data. After each cardiac arrest, the data were downloaded (by universal serial bus or bluetooth) onto a computer. For each event, the overall rSo2 for each patient was determined by calculating the mean of the continuous data (sampled every 4 s) from the time the sensor had been placed on the patient until either ROSC had been achieved or until CPR had been terminated as documented on the cardiac arrest records. ROSC was determined whenever (after a 2-minute period of CPR) a palpable pulse was elicited. The cerebral oximeter was then marked to indicate ROSC. We only analyzed the data from the start of CPR until that point. Artifact values were recognized by either an absent value or a single value obtained during a given 4-second period that was markedly different to the 4-second cycle either before or after it. There were thus 15 values routinely measured per minute, and any artifact was deleted. We did not have a warm-up or run-in period. The calculations of the mean value for each patient was carried out by A.N. and A.A. who were not blinded to ROSC outcome or mode of chest compression. Although staff were not specifically blinded to the cerebral oximetry values, the monitor was routinely placed away from the actual cardiac arrest and staff did not have any specific understanding regarding the interpretation of rSo2 values. Thus, values were not used for the purpose of clinical decision making. The decision to use either M-CPR or A-CPR was at the discretion of the cardiac arrest team. Patients in A-CPR and M-CPR groups were totally separated. At the time of this study, neither end-tidal Co2 (Petco2) monitoring nor quantitative methods aimed at determining the quality of chest compressions were routinely available on the hospital wards, and we were thus unable to compare data derived from cerebral oximetry with these resuscitation variables. Statistical Analysis Patients’ characteristics between those using M-CPR and A-CPR were compared using Fisher exact test or chi-square test for categorical variables and Wilcoxon rank-sum test for continuous variables. The mean rSo2 comparison between patients who achieved ROSC and those who did not achieve ROSC and also between patients who used A-CPR and those who used M-CPR was further assessed in a two-way analysis of variance model. Normality assumption was confirmed. Statistical significance level was set at 0.05, and p values from twosided test were reported. All analyses were performed in SAS 9.3 (SAS Institute, Cary, NC).
RESULTS In a 1-year period, a total of 34 patients were studied. Twentytwo underwent M-CPR and 12 underwent A-CPR. Patients were not randomly assigned and the decision to use either A-CPR or M-CPR was left at the discretion of the cardiac arrest team. In general, the decision to use oximetry and/or A-CPR seemed to reflect the overall level of familiarity with these www.ccmjournal.org
931
Parnia et al
technologies. They were used predominantly more toward the later stages of the clinical evaluation as staff felt more comfortable integrating these into their usual practice. Decisions were not made based on the team’s perception regarding a patients’ chance of survival. There were no significant differences in age, initial rhythm, duration of cardiac arrest, dose of epinephrine, or dose of atropine between patients receiving M-CPR and A-CPR (Table 1). In patients receiving cerebral oximetry monitoring, a significant difference was noted in the rSo2 in those who achieved ROSC compared with those without ROSC (median ± interquartile range: 47.4% ± 21.4% vs 23% ± 18.4%, p < 0.001) (Fig. 1A). Comparing the results between the two different CPR delivery methods, a significantly higher mean rSo2 was observed in the A-CPR group compared with the M-CPR group (53.1% ± 23.4% vs 24% ± 25%, p = 0.002) (Fig. 1B). There was still strong statistical evidence to support this finding after adjusting for patients who had achieved ROSC or not (p < 0.001).
DISCUSSION While acknowledging that our study has a number of significant inherent limitations which predominantly relate to the observational nature of the work and the use of a convenience sample of patients, our results suggest that a higher level of rSo2 may be associated with ROSC. Furthermore, the observation that in our hospital patients were more likely to achieve ROSC using A-CPR than M-CPR may reflect the challenges that exist in ensuring staff assigned to provide CPR can maintain consistent chest compressions, rather than a specific benefit with respect to the mode of CPR delivery. Because of the limitation in sample size, other possible confounders such as age and patient’s disease could not be considered to estimate the odds ratio of achieving ROSC using A-CPR or M-CPR. We Table 1.
further acknowledge that the patient’s underlying condition is of paramount importance in determining whether ROSC is achieved and we cannot exclude potential bias in this evaluation. Therefore, larger studies are needed to determine the relationship between rSo2 and ROSC, as well as the role of A-CPR and M-CPR in achieving higher levels of rSo2. Our results further suggest that cerebral oximetry may also provide information regarding the quality of oxygen delivery to the heart. This observation may simply reflect the reality that achieving higher levels of oxygen delivery and perfusion to the brain depends on the overall quality of the circulation, which will most likely impact other organs too. In previous studies, it has been shown that cerebral oximetry may potentially be useful as a predictor of survival and neurological outcomes following cardiac arrest (12–15). In one small study, it was demonstrated that a mean rSo2 of 17% or less achieved during out-of-hospital cardiac arrest appeared to predict nonsurvival (12). This was supported by a larger study in which patients arriving in the emergency department following out-of-hospital cardiac arrest with an rSo2 of less than or equal to 15% were found not to survive (14). Furthermore, it has been demonstrated that in a similar patient population, an rSo2 less than 25% on arrival to the emergency room corresponds with poor outcomes (15). The results of our study complement these findings but additionally indicate that rSo2 monitoring can potentially be used as a dynamic rather than a static measure during cardiac arrest. Thus, an effective intervention (such as standardizing the quality of chest compressions) may increase oxygen delivery as indicated by rSo2. Furthermore, the use of cerebral oximetry during cardiac arrest may compliment the use of Petco2 monitoring, which has been proposed as a marker of the quality of organ perfusion and ROSC (16). Petco2 monitoring is impacted by underlying pulmonary pathology and
Patient Demographics and Characteristics Automated Mechanical Chest Compression Device (n = 12)
Variables
Age (mean ± sd)
p
72.77 ± 15.27
0.2409
7 (58.33)
15 (68.18)
0.7113
Pulseless electrical activity
5 (41.67)
15 (68.18)
Asystole
4 (33.33)
5 (22.73)
Ventricular fibrillation/ventricular tachycardia
3 (25)
2 (9.09)
Male, %
67.67 ± 9.58
Manual Chest Compression Device (n = 22)
Initial rhythm, %
Duration (min) (mean ± sd)
0.3206
15.92 ± 10.87
20.64 ± 13.74
0.2812
Epinephrine dose (mg) (median ± IQR)
5.00 ± 5.00
5.00 ± 3.00
0.8257
Atropine dose (mg) (median ± IQR)
2.00 ± 2.00
1.00 ± 2.00
9.5 ± 3.00
4.5 ± 7.00
Time to start of oximetry measurements (min) (median ± IQR)
1 0.2596
IQR = interquartile range. p value is based on Wilcoxon rank-sum test for continuous variables and Fisher exact test or chi-square test for categorical variables.
932
www.ccmjournal.org
April 2014 • Volume 42 • Number 4
Neurologic Critical Care
REFERENCES
Figure 1. The relationship between cerebral oxygen saturation (rSo2) and return of spontaneous circulation (ROSC) following cardiac arrest (A) and the impact of automated mechanical chest compression (A-CPR) and conventional manual chest compression (M-CPR) on rSo2 (B). p values were based on Wilcoxon rank-sum test.
pulmonary edema during cardiac arrest, which may lead to readings that do not accurately reflect the quality of the circulation, whereas cerebral oximetry does not have these inherent limitations. Thus, further studies are needed to compare these technologies in their ability to predict ROSC and overall survival outcomes. Finally, interestingly there is evidence that a higher level of rSo2 may be required to improve longer term outcomes. In one small study of 18 out-of-hospital cardiac arrest patients who had achieved ROSC, it was demonstrated that patients who survived for 1 week (n = 9) had achieved a significantly higher median cerebral oxygenation during their resuscitation than those who did not survivor to 1 week (63% vs 46%, p = 0.003) (13). However, larger studies are now needed to determine an optimal cutoff of rSo2 level for clinical use. Further larger studies should also be aimed at identifying factors that can improve the quality of cerebral perfusion and oxygen delivery during cardiac arrest with a view to determining the optimal mean rSo2 that may potentially lead to improved long-term survival and neurological outcomes.
CONCLUSIONS The integration of cerebral oximetry into cardiac arrest resuscitation provides a novel noninvasive method to determine the quality of cerebral perfusion and oxygen delivery to the brain. The use of A-CPR device during in-hospital cardiac arrest may lead to improved oxygen delivery and organ perfusion.
Critical Care Medicine
1. Nadkarni VM, Larkin GL, Peberdy MA, et al; National Registry of Cardiopulmonary Resuscitation Investigators: First documented rhythm and clinical outcome from in-hospital cardiac arrest among children and adults. JAMA 2006; 295:50–57 2. Neumar RW, Nolan JP, Adrie C, et al: Post-cardiac arrest 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 Resuscitation 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; 118:2452–2483 3. Pollard V, Prough DS: Cerebral oxygenation: Near-infrared spectroscopy. In: Principles and Practice of Intensive Care Monitoring. Tobin MJ (Ed.) New York, McGraw-Hill, 1997, pp 1019–1035 4. Hightower D, Thomas SH, Stone CK, et al: Decay in quality of closedchest compressions over time. Ann Emerg Med 1995; 26:300–303 5. Wik L, Kramer-Johansen J, Myklebust H, et al: Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JAMA 2005; 293:299–304 6. Ashton A, McCluskey A, Gwinnutt CL, et al: Effect of rescuer fatigue on performance of continuous external chest compressions. Resuscitation 1998; 37:149–152 7. Parnia S, Nasir A, Shah C, et al: A feasibility study evaluating the role of cerebral oximetry in predicting return of spontaneous circulation in cardiac arrest. Resuscitation 2012; 83:282–985 8. Pollard V, Prough DS: Cerebral oxygenation: Near-infrared spectroscopy. In: Principles and Practice of Intensive Care Monitoring. Tobin MJ (Ed). New York, McGraw-Hill Professional, 1997, pp 1019–1024 9. Jöbsis FF: Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 1977; 198:1264–1267 10. Pollard V, Prough DS, DeMelo AE, et al: Validation in volunteers of a near-infrared spectroscope for monitoring brain oxygenation in vivo. Anesth Analg 1996; 82:269–277 11. McCormick PW, Stewart M, Ray P, et al: Measurement of regional cerebrovascular haemoglobin oxygen saturation in cats using optical spectroscopy. Neurol Res 1991; 13:65–70 12. Newman DH, Callaway CW, Greenwald IB, et al: Cerebral oximetry in out-of-hospital cardiac arrest: Standard CPR rarely provides detectable hemoglobin-oxygen saturation to the frontal cortex. Resuscitation 2004; 63:189–194 13. Müllner M, Sterz F, Binder M, et al: Near infrared spectroscopy during and after cardiac arrest—Preliminary results. Clin Intensive Care 1995; 6:107–111 14. Ito N, Nanto S, Nagao K, et al: Regional cerebral oxygen saturation predicts poor neurological outcome in patients with out-of-hospital cardiac arrest. Resuscitation 2010; 81:1736–1737 15. Ito N, Nanto S, Nagao K, et al: Regional cerebral oxygen saturation on hospital arrival is a potential novel predictor of neurological outcomes at hospital discharge in patients with out-of-hospital cardiac arrest. Resuscitation 2012; 83:46–50 16. Neumar RW, Otto CW, Link MS, et al: Part 8: Adult advanced cardiovascular life support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010; 122:S729–S767
www.ccmjournal.org
933
A Feasibility Study of Cerebral Oximetry During In-Hospital Mechanical and Manual Cardiopulmonary Resuscitation* Sam Parnia, MD, PhD; Asad Nasir, MD; Anna Ahn, MD; Hanan Malik, MD; Jie Yang, PhD; Jiawen Zhu, MS; Francis Dorazi, DO; Paul Richman, MD
Objective: A major hurdle limiting the ability to improve the quality of resuscitation has been the lack of a noninvasive real-time detection system capable of monitoring the quality of cerebral and other organ perfusion, as well as oxygen delivery during cardiopulmonary resuscitation. Here, we report on a novel system of cerebral perfusion targeted resuscitation. Design: An observational study evaluating the role of cerebral oximetry (Equanox; Nonin, Plymouth, MI, and Invos; Covidien, Mansfield, MA) as a real-time marker of cerebral perfusion and oxygen delivery together with the impact of an automated mechanical chest compression system (Life Stat; Michigan Instruments, Grand Rapids, MI) on oxygen delivery and return of spontaneous circulation following in-hospital cardiac arrest. Setting: Tertiary medical center. Patients: In-hospital cardiac arrest patients (n = 34). Main Results: Cerebral oximetry provided real-time information regarding the quality of perfusion and oxygen delivery. The use of automated mechanical chest compression device (n = 12) was associated with higher regional cerebral oxygen saturation compared with manual chest compression device (n = 22) (53.1% ± 23.4% vs 24% ± 25%, p = 0.002). There was a significant difference in mean regional cerebral oxygen saturation (median % ± interquartile range) in patients who achieved return of sponta-
*See also p. 1001. All authors: Resuscitation Research Group, Division of Pulmonary & Critical Care, Department of Medicine, State University of New York at Stony Brook, Stony Brook Medical Center, T17-040 Health Sciences Center, Stony Brook, NY. Supported, in part, by Stony Brook Medical Center, Department of Medicine Pilot Project Program Award. The authors have disclosed that they do not have any potential conflicts of interest. Address requests for reprints to: Sam Parnia, MD, PhD, MRCP, State University of New York at Stony Brook, Stony Brook Medical Center, T17040 Health Sciences Center, Stony Brook, NY 11794–8172. E-mail: sam. [email protected] Copyright © 2013 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0000000000000047
930
www.ccmjournal.org
neous circulation (n = 15) compared with those without return of spontaneous circulation (n = 19) (47.4% ± 21.4% vs 23% ± 18.42%, p < 0.001). After controlling for patients achieving return of spontaneous circulation or not, significantly higher mean regional cerebral oxygen saturation levels during cardiopulmonary resuscitation were observed in patients who were resuscitated using automated mechanical chest compression device (p < 0.001). Conclusions: The integration of cerebral oximetry into cardiac arrest resuscitation provides a novel noninvasive method to determine the quality of cerebral perfusion and oxygen delivery to the brain. The use of automated mechanical chest compression device during in-hospital cardiac arrest may lead to improved oxygen delivery and organ perfusion. (Crit Care Med 2014; 42:930–933) Key Words: cardiac arrest; cerebral oximetry; cerebral perfusion; near-infrared spectroscopy; resuscitation; return of spontaneous circulation
A
lthough in-hospital resuscitation leads to return of spontaneous circulation (ROSC) in up to 50%, survival to discharge remains poor at 18% (1). Pathophysiologically, cardiac arrest reflects an injury process that begins with the cessation of heartbeat and reflects a state of imbalanced oxygen delivery relative to tissue oxygen requirements, which if left unrecognized and untreated can lead to progressive and irreversible organ damage (2). Even though one of the major goals of cardiopulmonary resuscitation (CPR) is maintaining adequate cerebral perfusion and oxygen delivery, to date there have been no standard noninvasive “real-time” clinical methods available to determine the balance between oxygen delivery and requirements in the brain during cardiac arrest resuscitation. Although jugular venous oxygen saturation (Sjo2) monitoring can determine the balance between cerebral oxygen delivery and oxygen requirements, it is an invasive test, which greatly limits its practical application during cardiac arrest resuscitation. A noninvasive technology that has emerged and is not susceptible to motion artifact during cardiac arrest is cerebral oximetry using near-infrared spectroscopy (NIRS). April 2014 • Volume 42 • Number 4
Neurologic Critical Care
NIRS measures the fall in optical intensity, expressed as the change in optical density per centimeter of tissue, and when applied to the brain, it provides real-time continuous information regarding regional cerebral oxygen saturation (rSo2) in the frontal lobe, which reflects the balance between oxygen delivery and consumption in cerebral tissue (3). Although the quality of chest compression delivery is of prime importance during cardiac arrest resuscitation, many studies have demonstrated significant variability in the delivery of chest compressions in hospitals (4–6). Analysis of the quality of CPR using a quantitative system of CPR (Real CPR; Zoll, Chelmsford, MA) during simulation training at our institution demonstrated that staff were able to achieve the rate and depth recommended by the American Heart Association guidelines less than 50% of the time. However, the use of an automated mechanical chest compression (A-CPR) device (Life Stat; Michigan Instruments, MI) demonstrated the correct delivery of depth and rate 100% of the time. Thus, to standardize the delivery of chest compression across the hospital, we introduced A-CPR. Furthermore, building on an earlier feasibility study in which NIRS was incorporated into a small series of in-hospital manual cardiac arrest resuscitations at our institution (7), we compared the quality of cerebral oxygen delivery using A-CPR device with conventional manual chest compression (M-CPR) device at our specific institution. We hypothesized that standardizing the rate and depth of CPR delivery using A-CPR would lead to improved brain oxygen delivery. The primary outcome measure for this study was the mean rSo2 level during CPR in the A-CPR group compared with the M-CPR group.
MATERIALS AND METHODS Institutional review board approval was obtained for this study following modification to the clinical method used to resuscitate cardiac arrest patients at Stony Brook Medical Center. We conducted an observational study in a convenience sample of 34 adult patients (age > 21) who underwent in-hospital cardiac arrest resuscitation during a 1-year clinical evaluation period. This was set up to examine the utility of incorporating cerebral oximetry using NIRS (Equanox 7600; Nonin, Plymouth, MI, and Invos; Somanetics, Troy, MI) during cardiac arrest. Each sensor consists of an adhesive strip, a near-infrared light transmitter, and two sensors. The emitted light is scattered by tissues in two parabolic curves: one corresponding with hemoglobin saturation from the skin and skull and the other from the skin, skull, and frontal cortex. Using specific algorithms, the hemoglobin saturation in the frontal cortex is calculated (8–11). After a cardiac arrest was announced through the paging system in the hospital, the cerebral oximeter was taken by a predesignated member of staff from the medical ICU (MICU) to the site of cardiac arrest, where an on-call MICU physician and nurse trained in its use also attended. The placement of a single sensor on either the right or left lateral aspect of the forehead was considered sufficient for the purpose of measuring the overall quality of cerebral oxygen delivery, since Critical Care Medicine
cerebral blood flow during cardiac arrest is predominantly dependent on the quality of the circulation. The cerebral oximeter recorded continuous data. After each cardiac arrest, the data were downloaded (by universal serial bus or bluetooth) onto a computer. For each event, the overall rSo2 for each patient was determined by calculating the mean of the continuous data (sampled every 4 s) from the time the sensor had been placed on the patient until either ROSC had been achieved or until CPR had been terminated as documented on the cardiac arrest records. ROSC was determined whenever (after a 2-minute period of CPR) a palpable pulse was elicited. The cerebral oximeter was then marked to indicate ROSC. We only analyzed the data from the start of CPR until that point. Artifact values were recognized by either an absent value or a single value obtained during a given 4-second period that was markedly different to the 4-second cycle either before or after it. There were thus 15 values routinely measured per minute, and any artifact was deleted. We did not have a warm-up or run-in period. The calculations of the mean value for each patient was carried out by A.N. and A.A. who were not blinded to ROSC outcome or mode of chest compression. Although staff were not specifically blinded to the cerebral oximetry values, the monitor was routinely placed away from the actual cardiac arrest and staff did not have any specific understanding regarding the interpretation of rSo2 values. Thus, values were not used for the purpose of clinical decision making. The decision to use either M-CPR or A-CPR was at the discretion of the cardiac arrest team. Patients in A-CPR and M-CPR groups were totally separated. At the time of this study, neither end-tidal Co2 (Petco2) monitoring nor quantitative methods aimed at determining the quality of chest compressions were routinely available on the hospital wards, and we were thus unable to compare data derived from cerebral oximetry with these resuscitation variables. Statistical Analysis Patients’ characteristics between those using M-CPR and A-CPR were compared using Fisher exact test or chi-square test for categorical variables and Wilcoxon rank-sum test for continuous variables. The mean rSo2 comparison between patients who achieved ROSC and those who did not achieve ROSC and also between patients who used A-CPR and those who used M-CPR was further assessed in a two-way analysis of variance model. Normality assumption was confirmed. Statistical significance level was set at 0.05, and p values from twosided test were reported. All analyses were performed in SAS 9.3 (SAS Institute, Cary, NC).
RESULTS In a 1-year period, a total of 34 patients were studied. Twentytwo underwent M-CPR and 12 underwent A-CPR. Patients were not randomly assigned and the decision to use either A-CPR or M-CPR was left at the discretion of the cardiac arrest team. In general, the decision to use oximetry and/or A-CPR seemed to reflect the overall level of familiarity with these www.ccmjournal.org
931
Parnia et al
technologies. They were used predominantly more toward the later stages of the clinical evaluation as staff felt more comfortable integrating these into their usual practice. Decisions were not made based on the team’s perception regarding a patients’ chance of survival. There were no significant differences in age, initial rhythm, duration of cardiac arrest, dose of epinephrine, or dose of atropine between patients receiving M-CPR and A-CPR (Table 1). In patients receiving cerebral oximetry monitoring, a significant difference was noted in the rSo2 in those who achieved ROSC compared with those without ROSC (median ± interquartile range: 47.4% ± 21.4% vs 23% ± 18.4%, p < 0.001) (Fig. 1A). Comparing the results between the two different CPR delivery methods, a significantly higher mean rSo2 was observed in the A-CPR group compared with the M-CPR group (53.1% ± 23.4% vs 24% ± 25%, p = 0.002) (Fig. 1B). There was still strong statistical evidence to support this finding after adjusting for patients who had achieved ROSC or not (p < 0.001).
DISCUSSION While acknowledging that our study has a number of significant inherent limitations which predominantly relate to the observational nature of the work and the use of a convenience sample of patients, our results suggest that a higher level of rSo2 may be associated with ROSC. Furthermore, the observation that in our hospital patients were more likely to achieve ROSC using A-CPR than M-CPR may reflect the challenges that exist in ensuring staff assigned to provide CPR can maintain consistent chest compressions, rather than a specific benefit with respect to the mode of CPR delivery. Because of the limitation in sample size, other possible confounders such as age and patient’s disease could not be considered to estimate the odds ratio of achieving ROSC using A-CPR or M-CPR. We Table 1.
further acknowledge that the patient’s underlying condition is of paramount importance in determining whether ROSC is achieved and we cannot exclude potential bias in this evaluation. Therefore, larger studies are needed to determine the relationship between rSo2 and ROSC, as well as the role of A-CPR and M-CPR in achieving higher levels of rSo2. Our results further suggest that cerebral oximetry may also provide information regarding the quality of oxygen delivery to the heart. This observation may simply reflect the reality that achieving higher levels of oxygen delivery and perfusion to the brain depends on the overall quality of the circulation, which will most likely impact other organs too. In previous studies, it has been shown that cerebral oximetry may potentially be useful as a predictor of survival and neurological outcomes following cardiac arrest (12–15). In one small study, it was demonstrated that a mean rSo2 of 17% or less achieved during out-of-hospital cardiac arrest appeared to predict nonsurvival (12). This was supported by a larger study in which patients arriving in the emergency department following out-of-hospital cardiac arrest with an rSo2 of less than or equal to 15% were found not to survive (14). Furthermore, it has been demonstrated that in a similar patient population, an rSo2 less than 25% on arrival to the emergency room corresponds with poor outcomes (15). The results of our study complement these findings but additionally indicate that rSo2 monitoring can potentially be used as a dynamic rather than a static measure during cardiac arrest. Thus, an effective intervention (such as standardizing the quality of chest compressions) may increase oxygen delivery as indicated by rSo2. Furthermore, the use of cerebral oximetry during cardiac arrest may compliment the use of Petco2 monitoring, which has been proposed as a marker of the quality of organ perfusion and ROSC (16). Petco2 monitoring is impacted by underlying pulmonary pathology and
Patient Demographics and Characteristics Automated Mechanical Chest Compression Device (n = 12)
Variables
Age (mean ± sd)
p
72.77 ± 15.27
0.2409
7 (58.33)
15 (68.18)
0.7113
Pulseless electrical activity
5 (41.67)
15 (68.18)
Asystole
4 (33.33)
5 (22.73)
Ventricular fibrillation/ventricular tachycardia
3 (25)
2 (9.09)
Male, %
67.67 ± 9.58
Manual Chest Compression Device (n = 22)
Initial rhythm, %
Duration (min) (mean ± sd)
0.3206
15.92 ± 10.87
20.64 ± 13.74
0.2812
Epinephrine dose (mg) (median ± IQR)
5.00 ± 5.00
5.00 ± 3.00
0.8257
Atropine dose (mg) (median ± IQR)
2.00 ± 2.00
1.00 ± 2.00
9.5 ± 3.00
4.5 ± 7.00
Time to start of oximetry measurements (min) (median ± IQR)
1 0.2596
IQR = interquartile range. p value is based on Wilcoxon rank-sum test for continuous variables and Fisher exact test or chi-square test for categorical variables.
932
www.ccmjournal.org
April 2014 • Volume 42 • Number 4
Neurologic Critical Care
REFERENCES
Figure 1. The relationship between cerebral oxygen saturation (rSo2) and return of spontaneous circulation (ROSC) following cardiac arrest (A) and the impact of automated mechanical chest compression (A-CPR) and conventional manual chest compression (M-CPR) on rSo2 (B). p values were based on Wilcoxon rank-sum test.
pulmonary edema during cardiac arrest, which may lead to readings that do not accurately reflect the quality of the circulation, whereas cerebral oximetry does not have these inherent limitations. Thus, further studies are needed to compare these technologies in their ability to predict ROSC and overall survival outcomes. Finally, interestingly there is evidence that a higher level of rSo2 may be required to improve longer term outcomes. In one small study of 18 out-of-hospital cardiac arrest patients who had achieved ROSC, it was demonstrated that patients who survived for 1 week (n = 9) had achieved a significantly higher median cerebral oxygenation during their resuscitation than those who did not survivor to 1 week (63% vs 46%, p = 0.003) (13). However, larger studies are now needed to determine an optimal cutoff of rSo2 level for clinical use. Further larger studies should also be aimed at identifying factors that can improve the quality of cerebral perfusion and oxygen delivery during cardiac arrest with a view to determining the optimal mean rSo2 that may potentially lead to improved long-term survival and neurological outcomes.
CONCLUSIONS The integration of cerebral oximetry into cardiac arrest resuscitation provides a novel noninvasive method to determine the quality of cerebral perfusion and oxygen delivery to the brain. The use of A-CPR device during in-hospital cardiac arrest may lead to improved oxygen delivery and organ perfusion.
Critical Care Medicine
1. Nadkarni VM, Larkin GL, Peberdy MA, et al; National Registry of Cardiopulmonary Resuscitation Investigators: First documented rhythm and clinical outcome from in-hospital cardiac arrest among children and adults. JAMA 2006; 295:50–57 2. Neumar RW, Nolan JP, Adrie C, et al: Post-cardiac arrest 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 Resuscitation 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; 118:2452–2483 3. Pollard V, Prough DS: Cerebral oxygenation: Near-infrared spectroscopy. In: Principles and Practice of Intensive Care Monitoring. Tobin MJ (Ed.) New York, McGraw-Hill, 1997, pp 1019–1035 4. Hightower D, Thomas SH, Stone CK, et al: Decay in quality of closedchest compressions over time. Ann Emerg Med 1995; 26:300–303 5. Wik L, Kramer-Johansen J, Myklebust H, et al: Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JAMA 2005; 293:299–304 6. Ashton A, McCluskey A, Gwinnutt CL, et al: Effect of rescuer fatigue on performance of continuous external chest compressions. Resuscitation 1998; 37:149–152 7. Parnia S, Nasir A, Shah C, et al: A feasibility study evaluating the role of cerebral oximetry in predicting return of spontaneous circulation in cardiac arrest. Resuscitation 2012; 83:282–985 8. Pollard V, Prough DS: Cerebral oxygenation: Near-infrared spectroscopy. In: Principles and Practice of Intensive Care Monitoring. Tobin MJ (Ed). New York, McGraw-Hill Professional, 1997, pp 1019–1024 9. Jöbsis FF: Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 1977; 198:1264–1267 10. Pollard V, Prough DS, DeMelo AE, et al: Validation in volunteers of a near-infrared spectroscope for monitoring brain oxygenation in vivo. Anesth Analg 1996; 82:269–277 11. McCormick PW, Stewart M, Ray P, et al: Measurement of regional cerebrovascular haemoglobin oxygen saturation in cats using optical spectroscopy. Neurol Res 1991; 13:65–70 12. Newman DH, Callaway CW, Greenwald IB, et al: Cerebral oximetry in out-of-hospital cardiac arrest: Standard CPR rarely provides detectable hemoglobin-oxygen saturation to the frontal cortex. Resuscitation 2004; 63:189–194 13. Müllner M, Sterz F, Binder M, et al: Near infrared spectroscopy during and after cardiac arrest—Preliminary results. Clin Intensive Care 1995; 6:107–111 14. Ito N, Nanto S, Nagao K, et al: Regional cerebral oxygen saturation predicts poor neurological outcome in patients with out-of-hospital cardiac arrest. Resuscitation 2010; 81:1736–1737 15. Ito N, Nanto S, Nagao K, et al: Regional cerebral oxygen saturation on hospital arrival is a potential novel predictor of neurological outcomes at hospital discharge in patients with out-of-hospital cardiac arrest. Resuscitation 2012; 83:46–50 16. Neumar RW, Otto CW, Link MS, et al: Part 8: Adult advanced cardiovascular life support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010; 122:S729–S767
www.ccmjournal.org
933
