Article, Cardiology

On detection of spontaneous pulse by photoplethysmography in cardiopulmonary resuscitation

a b s t r a c t

Objective: This work investigates the potential of photoplethysmography (PPG) to detect a spontaneous pulse from the finger, nose or ear in order to support pulse checks during cardiopulmonary resuscitation (CPR).

Methods: In a prospective single-center cross-sectional study, PPG signals were acquired from cardiac arrest vic-

tims who underwent CPR. The PPG signals were analyzed and compared to arterial blood pressure (ABP) signals as a reference during three distranaisco; Date: 2/2/2020; Time:18:44:23inct phases of CPR: compression pauses, on-going compressions and at very low arterial blood pressure. Data analysis was based on a qualitative subjec- tive visual description of similarities of the frequency content of PPG and ABP waveform.

Results: In 9 patients PPG waveforms corresponded to ABP waveforms during normal blood pressures. During ABP in the clinically challenging range of 60 to 90 mmHg and during chest compressions and pauses, PPG contin- ued to resemble ABP, as both signals showed similar frequency components as a result of chest compressions as well as Cardiac activity. Altogether 1199 s of PPG data in compression pauses were expected to show a spontane- ous pulse, of which 732 s (61%) of data were artifact-free and showed the spontaneous pulse as visible in the ABP. Conclusions: PPG signals at all investigated sites can indicate pulse presence at the moment the heart resumes beating as verified via the ABP signal. Therefore, PPG may provide decision support during CPR, especially related to preventing and shortening interruptions for unnecessary pulse checks. This could have impact on CPR out- come and should further be investigated.

(C) 2019 The Authors. This is an open access article under the CC BY-NC-ND license (http://

creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Objectified assessment of patient condition during cardiopulmonary resuscitation (CPR), especially assessment of cardiac arrest and return of spontaneous circulation (ROSC) during chest compressions, can be very challenging and is often not possible, yet essential for treatment

Abbreviations: ABP, arterial blood pressure; CA, cardiac arrest; CPR, cardiopulmonary resuscitation; ECG, electrocardiography; ETCO2, end tidal CO2; IQR, interquartile range; PEA, pulseless electrical activity; PPG, photo plethysmography; ROSC, return of spontaneous circulation; VF, ventricular fibrillation.

* Corresponding author at: Universitatsklinik fur Notfallmedizin, Medizinische Universitat Wien, Allgemeines Krankenhaus der Stadt Wien, Wahringer Gurtel 18-20/ 6D, 1090 Wien, Austria.

E-mail addresses: [email protected] (P. Hubner), [email protected] (R.W.C.G.R. Wijshoff), [email protected] (J. Muehlsteff), [email protected] (C. Wallmuller), [email protected] (A.M. Warenits),

[email protected] (I.A.M. Magnet), [email protected] (K. Nammi), [email protected] (F. Sterz).

decisions. It is expected that additional physiological information may be useful in adjusting CPR to the state of the patient, but there has been little practical progress towards this goal.

Manual pulse check during CPR, whether performed by first re- sponders or professional helpers, is difficult to perform, often unreliable and requires interruption of chest compressions [1-3]. If too long, these interruptions can negatively impact outcome [4,5] and therefore mini- mizing pauses is strongly recommended [6]. In CPR guidelines, recom- mendation for pulse check by lay responders was removed and only conditional approval for pulse check by professional responders was retained [7]. A technical solution providing a quick and reliable pulse check could permit re-introduction of this important assessment, but has yet to be forthcoming [8].

Photoplethysmography (PPG) is the optical technology used in Pulse oximeters. PPG measures changes in blood volume by emitting light through tissue, is used to monitor spontaneous pulse rate and other he- modynamic parameters [9-11] and is a non-invasive and easy-to-use technology. It is ubiquitously applied in Acute care settings for patient

https://doi.org/10.1016/j.ajem.2019.05.044

0735-6757/(C) 2019 The Authors. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

monitoring. However, only recently technical progress in signal fusion and signal processing made PPG-based pulse detection during periodi- cal motion feasible. This technology is now in widespread use in health watches for fitness consumer applications. Besides this technical prog- ress we showed in pre-clinical CPR studies under controlled conditions [12,13] that PPG appears to have the potential to indicate when the heart resumes beating and should therefore be reconsidered as a tool in CPR decision support.

The goal of this work was to investigate the potential of PPG to detect the presence and rate of spontaneous pulse during chest compressions, pauses in compressions and after ROSC in a clinical setting.

Materials and methods

Hemodynamic waveforms and supporting clinical data of 19 pa- tients undergoing CPR with chest compressions were collected in a pro- spective single-center cross-sectional study, conducted at the Department of Emergency Medicine, Medical University Vienna, from October 2012 until January 2014.

Ethics, consent and permissions

The investigation complied with the Declaration of Helsinki’s princi- ples for physicians engaged in biomedical research involving human subjects and was approved by the appropriate ethics committee. Since enrolled patients were all unconscious, in cardiac arrest and undergoing CPR, the Institutional Review Board waived the need for informed con- sent. [https://ekmeduniwien.at/core/catalog/2012 (EK-Nr: 1574/ 2012)].

Setup and patients

The Department of Emergency Medicine is an interdisciplinary emergency department (ED) with a complete 14-bed intensive care unit (ICU), where patients are routinely treated after resuscitation. Most patients in this study were brought to our ICU by the ambulance service after suffering out-of-hospital cardiac arrest. Some patients suf- fered cardiac arrest at the ED during workup. Patients fit inclusion criteria if they presented at the ICU of the ED in cardiac arrest and underwent CPR. If they fit inclusion criteria and enrolment in this study was assumed not to interfere with their routine medical care, non-invasive PPG sensors were attached to finger, nose and ear, as de- scribed in detail below. No patient had to be excluded from the study because of interference with routine medical care.

Standard monitors

Arterial blood pressure (ABP), Electrocardiography and pulse oximetry data were collected from Philips Intellivue(R) (Philips GmbH, Vienna, Austria) bedside monitors. Arterial blood pressure was mea- sured from a catheter in the Radial artery. Pulse oximetry was gathered from a finger. The pulse oximetry module provided a filtered PPG signal, referred to as plethwave.

Supplemental monitors

A Philips Respironics(R) NICO Cardiopulmonary Management System monitor (Philips GmbH, Vienna, Austria) was used for capnography and a second finger pulse oximetry measurement at the opposite hand from that used by the Intellivue pulse oximeter. Philips Novametrix Oxypleth(R) Pulse oximetry monitors (Philips GmbH, Vienna, Austria) measured data at the nose (nasal septum or alar wing) and the ear pinna, with standard “Y sensors” and ear clips. These monitors provided unfiltered, “raw”, red and infrared PPG signals.

Data collection

Data from the Intellivue, NICO and Oxypleth monitors was recorded on a dedicated Laptop computer using custom software. Waveforms originating from different monitors were synchronized using custom software. Additionally, clinical data including demographic information, clinical circumstances and Short-term outcome, as well as blood gas Lab- oratory results was gathered for each patient.

Analysis

PPG signals were studied during three distinct phases in CPR: during compression pauses, during on-going compressions and after ROSC at very low ABP. The PPG signal time-traces as well as their frequency con- tent were studied. The frequency content of the signals was displayed via spectrograms, which visualize which frequencies (oscillating com- ponents) are present in the signal over time. In this prospective single-center cross-sectional study our primary objective was limited to a qualitative visual description of similarities between PPG and ABP waveforms. We defined similarity between PPG and ABP differently for the three distinct phases in CPR. During pauses and after ROSC at very low ABP, we defined similarity as both PPG and ABP signals show- ing either presence or absence of a spontaneous pulse. During compres- sions, we defined similarity as both PPG and ABP signals showing either only quasi-periodic compression artifacts, or a more complex shape be- cause a compression artifact as well as a spontaneous pulse were pres- ent in the signal, as we observed in preclinical studies [13]. Overall results were presented for the PPG signals measured in the compression pauses by visually determining for what percentage of time the PPG sig- nals showed a spontaneous pulse, when this was expected based on the ABP reference signal. In this preliminary investigation with its small sample size, and with the current clinical evidence at hand, we thought this presentation to be most appropriate. The goal was limited to show- ing the potential of PPG to detect spontaneous pulse during CPR, such as during compression pauses and during on-going compressions.

Results

Data from 9 of 19 patients was sufficient for analysis. Table 1 pro- vides demographic details on these patients.

Description of PPG signals

Assessment of PPG signals in compression pauses

Fig. 1 shows the PPG signals from finger, nose and ear compared to the reference signals ABP, ECG and EtCO2 during phases with and with- out chest compressions, acquired from patient 3. Artifacts in the

Table 1

Patients’ demographics.

Patients n = 9

Age

61 years (IQR 34-70)

Female sex

1

Witnessed

8

OOH

6

VF

2

PEA

6

Asystole

1

ROSC

9

Time to ROSC

55 min (IQR 1-108)

Cardiac Cause

5

Pulmonary Cause

1

Cerebral Cause

1

Sepsis

1

Drowning

1

IQR indicates interquartile range; OOH out of hospital; VF ventricular fibrillation; PEA pulseless electrical activity; ROSC return of spontaneous circulation.

Fig. 1. PPG signals from finger, nose and ear (top four traces), with ABP (trace five), ECG (trace six) and EtCO2 (trace seven) as references, acquired from patient 3. When chest compressions stop, as marked by the dashed line, the two peripheral finger PPG signals directly show presence of a spontaneous pulse, whereas the nose and ear PPG signals show a spontaneous pulse with delay, as indicated by the arrows. Artifacts in the waveforms are marked based on visual inspection. ABP: arterial blood pressure; ECG: electrocardiography; EtCO2: end-tidal CO2; IR: infrared; PPG: photoplethysmography.

waveforms are marked based on visual inspection. After compressions stopped, as indicated by the dashed vertical line, systolic blood pressure exceeded 60 mmHg, a stabilizing ECG rhythm was observed and EtCO2

built up over 40 mmHg, which was indicative of ROSC for this patient. In the period without compressions, spontaneous pulses were visible in all PPG signals coinciding with the spontaneous pulses in the ABP wave.

Fig. 2. PPG signals from finger, nose and ear (top three traces), with ABP (trace four) and EtCO2 (trace five) as references, acquired from patient 5 during ECLS with non-pulsatile flow. When a spontaneous pulse develops in the ABP signal, pulses directly appear in the central nose and ear PPG signals and with a delay in the finger PPG signal, as indicated by the arrows. Artifacts in the waveforms are marked based on visual inspection. ABP: arterial blood pressure; ECLS: Extracorporeal life support; EtCO2: end-tidal CO2; IR: infrared; PPG: photoplethysmography.

After compressions stopped, the spontaneous pulses appeared in the PPG signals with different delays, as indicated by the arrows. In both pe- ripheral finger PPG signals, the spontaneous pulses appeared immedi- ately after compressions stopped, but in the central nose and ear PPG signals, spontaneous pulses appeared after a delay of about 10 and 20 s, respectively. Furthermore, pulse morphology was different in the processed plethwave acquired from the finger (IntelliVue monitor) and the raw PPG signal acquired from the finger (NICO monitor).

Fig. 2 shows the PPG signals from finger, nose and ear with as refer- ence signals the ABP and EtCO2 signals, acquired from patient 5. Fig. 2 shows the episode where patient 5 undergoes extracorporeal life sup- port (ECLS) with a non-pulsatile flow and develops a spontaneous pulse again. Prior to ECLS, patient 5 first underwent CPR with compres- sions, which was unsuccessful and which episode precedes the ECLS ep- isode shown in Fig. 2. Artifacts in the waveforms in Fig. 2 are marked based on visual inspection. We see that pulses appeared in all PPG sig- nals once the heart resumed beating. As indicated by the arrows, the spontaneous pulses appeared directly in the central nose and ear PPG signals, but were delayed by about 50 s in the finger PPG signal. Pulses appear in the central PPG signals at Pulse pressures as low as only 4 mmHg. All PPG signals showed spontaneous pulses before EtCO2 exceeded 40 mmHg, which can be considered indicative of ROSC. Also here pulse morphology was different in the processed plethwave ac- quired from the finger (IntelliVue monitor) and the raw PPG signal ac- quired from the finger (NICO monitor).

Fig. 3 shows PPG signals from finger, nose and ear, with as references ABP and EtCO2, acquired from patient 1 during a sequence of compres- sions and no compressions. During compressions all acquired waves showed compression-induced artifacts. In the period without compres- sions, a spontaneous pulse was visible in the ABP wave with a low mean pressure of about 19 mmHg and a small pulse pressure of about 18 mmHg. At that time, EtCO2 was about 40 mmHg. Consequently,

there was no clear indication of ROSC. In this period without compres- sions, only the nose PPG signal showed presence of a spontaneous pulse. Fig. 4 shows PPG signals from finger, nose and ear, with as references ABP and ECG, acquired from patient 9 during a sequence of compres- sions and no compressions. During compressions all acquired waves showed compression-induced artifacts. In the period without compres- sions, a spontaneous pulse was visible in the ABP wave with a mean pressure of about 90 mmHg and a small pulse pressure of about 5 mmHg. In this period without compressions, only the finger PPG sig-

nal showed presence of a spontaneous pulse.

Table 2 shows for what percentage of time the PPG signals measured during compression pauses showed a spontaneous pulse, when this was to be expected based on the ABP reference signal. The expected time per PPG signal indicates the maximum time that one could expect to ob- serve a spontaneous pulse in the PPG signal. We defined this maximum as the time that the ABP shows a spontaneous pulse and there were no coinciding artifacts in the PPG signal. The maximum time can therefore differ for each individual PPG signal. Artifacts have been identified based on morphology (e.g., spikes, irregular/non-consistent appearance, large pulsatility) and have been considered to indicate attachment issues or motion of the sensor, and were therefore excluded. Spontaneous pulses in the ABP signal have been identified based on visual inspection. Ar- rhythmic pulses have been included as well as small pulses with pulse pressures of 5-10 mmHg. Patient 5 first underwent CPR with compres- sions, but did not develop a spontaneous pulse until the CPR procedure was changed to ECLS. As there was no spontaneous pulse during CPR with compressions, Table 2 lists “n.a.” for patient 5.

When looking at the data from the finger sensor of the Intellivue monitor, which is routinely used at the emergency department, the PPG signal was expected to show a spontaneous pulse for at most 161 s, while it showed a spontaneous pulse for 81 s (50%). The NICO fin- ger PPG signal was expected to show a spontaneous pulse for 307 s,

Fig. 3. PPG signals from finger, nose and ear (top four traces), with ABP (trace five) and EtCO2 (trace six), acquired from patient 1. During the pause in chest compressions in between the two dashed lines, blood pressure is low and only the nose PPG signal shows presence of a spontaneous pulse. Artifacts in the waveforms are marked based on visual inspection. ABP: arterial blood pressure; EtCO2: end-tidal CO2; IR: infrared; PPG: photoplethysmography.

Fig. 4. PPG signals from finger, nose and ear (top three traces), with ABP (trace four) and ECG (trace five), acquired from patient 9. During the pause in chest compressions, as indicated by the blue bar in between the dashed lines, only the finger PPG signal shows presence of a spontaneous pulse. Artifacts in the waveforms are marked based on visual inspection. ABP: arterial blood pressure; ECG: electrocardiography; IR: infrared; PPG: photoplethysmography

while it showed a spontaneous pulse for 199 s (65%). The nose PPG sig- nal was expected to show a spontaneous pulse for 326 s, while it showed a spontaneous pulse for 244 s (75%). And the ear PPG signal was expected to show a spontaneous pulse for 405 s, while it showed a spontaneous pulse for 208 s (51%).

Pulse Detection during Compressions. Fig. 5 shows finger, nose and ear PPG signals with ABP and ECG sig-

nals as physiological references, acquired from patient 6 when the heart resumes beating during on-going chest compressions. When compres- sions were delivered during cardiac arrest, to the left of the first vertical dashed line, the PPG signals showed compression artifacts with a rela- tively stable morphology. When the heart resumed beating during com- pressions, as the rising blood pressure indicated in between the dashed vertical lines, the morphology of the PPG signals suddenly changed, the onset of which is marked by the arrows. Here, the PPG signal started to look more complex, as a result of the appearance of a spontaneous pulse component in the PPG signal, which is superimposed to the chest

compression artifacts. When the compressions stopped to the right of the second dashed vertical line, all PPG signals kept showing spontane- ous pulses as can be verified from the ABP and ECG signals. The pulsatility of the spontaneous pulse varied among the PPG signals, with the finger PPG signal having the weakest pulsatility and the nose PPG signal having the strongest pulsatility. After compressions stopped, the levels in the ABP were indicative of ROSC.

Fig. 6 presents the spectrograms of the finger, nose and ear PPG sig- nals and the ABP signal of patient 6, of which the corresponding time traces are shown in Fig. 5. Spectrograms depict what frequencies (oscil- lating components) are present in a signal over time by means of a color-coding. The darker red the color, the stronger this frequency com- ponent is in the signal. All spectrograms showed a component at the chest compression rate of about 100 min-1. When compressions stopped at about 24:40, as marked by the dashed vertical line, the chest compression component disappeared in all spectrograms. At about 23:40 a second frequency component showed up in all

Table 2

Analysis of periods without compressions with an arterial blood pressure (ABP) – pulse and no photoplethysmography (PPG) – artifacts.

Patient Finger (Intellivue) Finger (NICO) Nose Ear

ABP, s

PPG, s (%)

ABP, s

PPG, s (%)

ABP, s

PPG, s (%)

ABP, s

PPG, s (%)

1

39

0 (0%)

38

0 (0%)

51

48 (94%)

49

8 (16%)

2

n.a.

n.a.

30

30 (100%)

21

21 (100%)

19

19 (100%)

3

38

24 (63%)

45

42 (93%)

45

24 (53%)

40

8 (20%)

4

36

9 (25%)

53

13 (25%)

n.a.

n.a.

66

0

5

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

6

n.a.

n.a.

61

34 (56%)

69

65 (94%)

63

54 (86%)

7

n.a.

n.a.

n.a.

n.a.

99

86 (87%)

92

86 (93%)

8

48

48 (100%)

n.a.

n.a.

n.a.

n.a.

28

12 (43%)

9

n.a.

n.a.

80

80 (100%)

41

0

48

21 (44%)

Total

161

81 (50%)

307

199 (65%)

326

244 (75%)

405

208 (51%)

ABP indicates the time per PPG signal that the ABP shows a spontaneous pulse and there is no artifact in the PPG signal – this is the maximum time that a spontaneous pulse could be expected in the PPG signal; s indicates seconds, % percent; n.a. not available.

Fig. 5. Finger, nose and ear PPG signals (top three traces), with ABP (trace four) and ECG (trace five) as references, acquired from patient 6. The change in morphology of the PPG signals during compressions, marked by the arrows between the dashed vertical lines, indicates presence of an underlying spontaneous pulse, as shown by the simultaneous rise in the ABP. When compressions stop, as marked by the right dashed vertical line, all PPG signals continue to show a spontaneous pulse. ABP: arterial blood pressure; ECG: electrocardiography; IR: infrared; PPG: photoplethysmography.

spectrograms, which was different than the compression rate. This was the spontaneous pulse rate of the patient, which varied over time during chest compressions and continues when chest compressions stopped. This spontaneous pulse component was what changed the morphology of the PPG waveforms in Fig. 5. As Fig. 6 illustrates, a spectrogram can be used to identify a spontaneous pulse component with a stable rate and to distinguish between the spontaneous pulse rate and the chest com- pression rate.

Discussion

Using state-of-the-art PPG sensors in 9 patients during CPR, PPG- based pulse detection at finger, nose and ear was feasible during ongo- ing compressions and compression pauses, confirming our pre-clinical findings [12-14]. Since measurements in pre-clinical studies were

made under controlled conditions, this data was not corrupted by arti- facts other than those resulting from compressions. However, since in reality conditions during CPR are different and much more difficult, pre-clinical studies stated the necessity of a clinical study to explore the use of PPG under actual conditions. In our study, we were able to il- lustrate that PPG, as additional method of monitoring patients in cardiac arrest, could provide pulse detection as a means for decision support in CPR. Placing arterial catheters in patients in cardiac arrest can be ex- tremely challenging, time consuming and is sometimes not possible until ROSC is achieved. We believe that especially patients not yet mon- itored by ABP could benefit a lot from this method of non-invasive mon- itoring. By showing absence of a spontaneous pulse, PPG could help prevent or shorten interruptions of chest compressions for unnecessary pulse checks. By showing presence of a spontaneous pulse during com- pressions, PPG may guide frequency and timing of stops in

Fig. 6. Spectrograms of the three PPG signals and ABP signal presented in Fig. 5, which show what frequencies are present in these signals over time. All spectrograms indicate the chest compression rate at about 100 min-1, which disappears when chest compressions stop at the dashed vertical line. At about 23:40, a second frequency component appears in all spectrograms, which is different than the compression rate. This is the spontaneous pulse rate of the patient, which continues when chest compressions stop at the dashed vertical line. ABP: arterial blood pressure; CPR: cardiopulmonary resuscitation; IR: infrared; PPG: photoplethysmography.

compressions and thereby help reduce the chance of compression- induced refibrillation. Another benefit of showing presence of a sponta- neous pulse might be that PPG could indicate to delay administration of a vasopressor and thereby reduce the chance of causing hemodynamic instability. The use of time-frequency representations for the PPG sig- nals (Fig. 6) is an important aspect of the analysis performed in this study. During CPR, the spectrogram helps in an intuitive way to identify the sudden appearance of a pulse signal caused by a spontaneously beating heart, and to distinguish this signal component from signal components due to compressions. Moreover, the signal components due to compressions can be easily identified in the spectrograms be- cause the compression-related frequencies can typically be derived from the thoracic impedance signal which is measured between the de- fibrillation pads. We also found that even standard finger PPG signals can show spontaneous pulses despite of expected centralization in some cases, which could be due to local vasodilation as a result of CO2 build-up. However, it seems reasonable that central-site sensors are preferred, since they suffer less from centralization.

Many studies have shown that pulse checks by manual palpation during resuscitation efforts are difficult to perform, are time- consuming and unreliable [1-3]. Therefore, various methods have been investigated for detection of ROSC during resuscitation, each with strengths and weaknesses [15-22]. The most reliable way to mon- itor patients under CPR is by measuring arterial blood pressure [23,24], however, this requires invasive procedures, which usually are not avail- able out-of-hospital and not immediately available in-hospital.

PPG is a ubiquitously applied method for monitoring patients in acute care settings [10-13]. However, as of today, it is not common clin- ical practice to use PPG to detect spontaneous pulse during CPR due to concerns about motion artifacts and poor Peripheral perfusion [25,26]. Interpretation of the PPG signal can be compromised by many factors like temperature, low perfusion and centralization of the patient or de- lays by signal processing and motion artifacts [26,27]. It may also be dif- ficult to determine if apparent pulses in the PPG signal are a result of chest compressions or of spontaneous cardiac activity [26]. Nonetheless, due to progress in hardware and signal processing techniques, PPG regained attention for pulse measurements even under severe motion, e.g., for pulse rate measurements in fitness applications with sensors in- tegrated in watches [28,29]. Additionally, there already is attention for using PPG specifically during CPR, e.g. in U.S. Patent 7,569,018 B [30], which describes providing audible feedback during CPR on the presence and amplitude of a spontaneous pulse in a PPG signal. This patent also mentions the benefit of indicating absence of a spontaneous pulse to prevent interruptions in CPR, specifically during PEA.

It should be emphasized that PPG can provide information on pres- ence/absence of a spontaneous pulse, but cannot detect ROSC as this re- quires assessing whether the circulation is life-sustaining. Detection of ROSC remains a clinical situational assessment. Furthermore, while ABP represents the blood pressure in the macrovasculature [10,11], PPG reflects the blood volume pulse in the microvasculature. This means that a single PPG signal cannot be used to assess the blood pres- sure of the patient, but only indicate whether a spontaneous pulse is present or not.

Limitations

The goal of this study was to determine the feasibility of identifying a spontaneous pulse in a PPG signal during CPR, not finding a technical so- lution to avoid motion artifacts. However, due to technical difficulties, especially the great vulnerability to artifacts, we were only able to use data from 9 patients for analysis from the original 19 patients in cardiac arrest and under CPR. Data from patients with one or another missing essential waveform during the periods of interest was excluded. Many artifacts and missing data were caused by the design of the PPG clips, which needs to be optimized for further research.

Spontaneous pulses can be detected in PPG signals, recorded from the finger, nose or ear, during short pauses in compressions, right after compressions stopped, or during ongoing compressions. PPG can furthermore measure a spontaneous pulse at very low ABP. There is in- sufficient data to conclude on the optimal site for PPG measurements during CPR in terms of detectability of a pulse at low perfusion or vasoconstricted states and in terms of vulnerability to artifacts. Vulner- ability to artifacts might be handled by a better PPG sensor design. Our findings indicate that PPG could potentially provide decision support during CPR and should be re-evaluated for this purpose.

Declarations of public or private research funding sources or foundations

In order to be able to conduct this study at our institution, it received support in the form of a grant and the equipment used from Philips Healthcare, Bothell, WA, USA.

Declaration of Competing Interest

Jens Muehlsteff and Ralph Wijshoff are employed by Philips Re- search, Patient Care & Measurements Group, Eindhoven, The Netherlands. Krishnakant Nammi is employed by Philips Healthcare, Bothell, WA, USA.

References

  1. Eberle B, Dick WF, Schneider T, Wisser G, Doetsch S, Tzanova I. Checking the carotid pulse check: diagnostic accuracy of First responders in patients with and without a pulse. Resuscitation 1996;33:107-16.
  2. Moule P. Checking the carotid pulse: diagnostic accuracy in students of the healthcare professions. Resuscitation 2000;44:195-201.
  3. Tibballs J, Russel P. Reliability of pulse palpation by Healthcare personnel to diagnose paediatric cardiac arrest. Resuscitation 2009;80(1):61-4 Jan.
  4. Wik L, Kramer-Johansen J, Myklebust H, Sorebo H, Svensson L, Fellows B, et al. Qual- ity of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JAMA 2005;293:299-304.
  5. Christenson J, Andrusiek D, Everson-Stewart S, Kudenchuk P, Hostler D, Powell J, et al. Berg R; Resuscitation Outcomes Consortium investigators. Chest compression fraction determines survival in patients with out-of-hospital ventricular fibrillation. Circulation 2009;120(13):1241-7 Sep 29.
  6. Neumar RW, Otto CW, Link MS, Kronick SL, Shuster M, Callaway CW, et al. Part 8: adult advanced cardiovascular life support: 2010 American Heart Association guide- lines for cardiopulmonary resuscitation and emergency cardiovascular care. Circula- tion 2010;122(18 Suppl 3):S729-67 Nov 2.
  7. Berg RA, Hemphill R, Abella BS, Aufderheide TP, Cave DM, Hazinski MF, et al. Part 5: adult basic life support: 2010 American Heart Association guidelines for cardiopul- monary resuscitation and emergency cardiovascular care. Circulation 2010;122(18 Suppl 3):S685-705 Nov 2.
  8. Rea TD, Cook AJ, Stiell IG, Powell J, Bigham B, Callaway CW, et al. Resuscitation out- comes consortium investigators. Predicting survival after out-of-hospital cardiac ar- rest: role of the Utstein data elements. Ann Emerg Med 2010;55(3):249-57 Mar.
  9. Shelley KH. Photoplethysmography: beyond the calculation of arterial oxygen satu- ration and heart rate. Anesth Analg 2007 Dec;105(6 Suppl):S31-6 [tables of contents].
  10. Reisner A, Shaltis PA, McCombie D, Asada HH. Utility of the photoplethysmogram in circulatory monitoring. Anesthesiology 2008;108(5):950-8 May.
  11. Allen J. Photoplethysmography and its application in clinical physiological measure- ment. Physiol Meas 2007;28(3):R1-39 Mar.
  12. Wijshoff RW, Mischi M, Woerlee PH, Aarts RM. Improving pulse oximetry accuracy by removing motion artifacts from photoplethysmograms using relative sensor mo- tion: a preliminary study. Adv Exp Med Biol 2013;789:411-7.
  13. Wijshoff RW, van der Sar T, Peeters WH, Bezemer R, Aelen P, Paulussen IW, et al. De- tection of a spontaneous pulse in photoplethysmograms during automated cardio- pulmonary resuscitation in a porcine model. Resuscitation 2013;84(11):1625-32 Nov.
  14. Wijshoff RW, van Asten AM, Peeters WH, Bezemer R, Noordergraaf GJ, Mischi M, et al. Photoplethysmography-based algorithm for detection of cardiogenic output during cardiopulmonary resuscitation. IEEE Trans Biomed Eng 2015;62(3):909-21 Mar.
  15. Pokorna M, Necas E, Kratochvil J, Skripsky R, Andrlik M, Franek O. A sudden increase in partial pressure end-tidal carbon dioxide (P(ET)CO(2)) at the moment of return of spontaneous circulation. J Emerg Med 2010;38(5):614-21 Jun.
  16. Grmec S, Krizmaric M, Mally S, Kozelj A, Spindler M, Lesnik B. Utstein style analysis of out-of-hospital cardiac arrest–bystander CPR and end expired carbon dioxide. Resuscitation 2007;72(3):404-14 Mar.
  17. Heradstveit BE, Sunde K, Sunde GA, Wentzel-Larsen T, Heltne JK. Factors complicat- ing interpretation of capnography during advanced life support in cardiac arrest–a clinical retrospective study in 575 patients. Resuscitation 2012;83(7):813-8 Jul.
  18. Risdal M, Aase SO, Kramer-Johansen J, Eftestol T. Automatic identification of return of spontaneous circulation during cardiopulmonary resuscitation. IEEE Trans Biomed Eng 2008;55(1):60-8 Jan.
  19. Stecher FS, Olsen JA, Stickney RE, Wik L. Transthoracic impedance used to evaluate performance of cardiopulmonary resuscitation during out of hospital cardiac arrest. Resuscitation 2008;79(3):432-7 Dec.
  20. Losert H, Risdal M, Sterz F, Nysaether J, Kohler K, Eftestol T, et al. Thoracic- impedance changes measured via defibrillator pads can monitor signs of circulation. Resuscitation 2007;73(2):221-8 May.
  21. Kamarainen A, Sainio M, Olkkola KT, Huhtala H, Tenhunen J, Hoppu S. Quality con- trolled manual chest compressions and cerebral oxygenation during in-hospital car- diac arrest. Resuscitation 2012;83(1):138-42 Jan.
  22. Reynolds JC, Salcido D, Koller AC, Sundermann ML, Frisch A, Suffoletto BP, et al. Tis- sue oximetry by Near-infrared spectroscopy in a porcine model of out-of-hospital cardiac arrest and resuscitation. Resuscitation 2013;84(6):843-7 Jun.
  23. Rivers EP, Lozon J, Enriquez E, Havstad SV, Martin GB, Lewandowski CA, et al. Simul- taneous radial, femoral, and aortic arterial pressures during human cardiopulmo- nary resuscitation. Crit Care Med 1993;21(6):878-83 Jun.
  24. Paradis NA, Martin GB, Rivers EP, Goetting MG, Appleton TJ, Feingold M, et al. Coro- nary perfusion pressure and the return of spontaneous circulation in human cardio- pulmonary resuscitation. JAMA 1990;263(8):1106-13 Feb 23.
  25. Xu J, Zhu H, Wang Z, Yu X, Walline J. Why do not we use finger pulse oximeter ple- thysmograph waveform to monitor the effectiveness of cardiopulmonary resuscita- tion? Resuscitation 2011;82(7):959 Jul.
  26. Moorthy SS, Dierdorf SF, Schmidt SI. Erroneous pulse oximeter data during CPR. Anesth Analg 1990;70(3):339 Mar.
  27. Spittal MJ. Evaluation of pulse oximetry during cardiopulmonary resuscitation. An- aesthesia 1993;48(8):701-3 Aug.
  28. Parak J, Tarniceriu A, Renevey P, Bertschi M, Delgado-Gonzalo R, Korhonen I. Evalu- ation of the beat-to-beat detection accuracy of PulseOn wearable optical heart rate monitor. Conf Proc IEEE Eng Med Biol Soc 2015;2015:8099-102 Aug.
  29. Tautan AM, Young A, Wentink E, Wieringa F. Characterization and reduction of mo- tion artifacts in photoplethysmographic signals from a wrist-worn device. Conf Proc IEEE Eng Med Biol Soc 2015;2015:6146-9.
  30. U.S. Patent No. 7,569,018 B1 (issued Aug. 4, 2009).

Leave a Reply

Your email address will not be published. Required fields are marked *