Data for: Reliability of mechanical ventilation during continuous chest compressions: A crossover study of transport ventilators in a human cadaver model of CPR

Abstract

The data presented in this article relate to the research article, “Reliability of mechanical ventilation during continuous chest compressions: a crossover study of transport ventilators in a human cadaver model of CPR” [1].

This article contains raw data of continuous recordings of airflow, airway and esophageal pressure during the whole experiment. Data of mechanical ventilation was obtained under ongoing chest compressions and from repetitive measurements of pressure-volume curves. All signals are presented as raw time series data with a sample rate of 200Hz for flow and 500 Hz for pressure. Additionally, we hereby publish extracted time series recordings of force and compression depth from the used automated chest compression device. Concomitantly, we report tables with time stamps from our laboratory book by which the data can be sequenced into different phases of the study protocol.

We also present a dataset of derived volumes which was used for statistical analysis in our research article together with the used exclusion list.

The reported dataset can help to understand mechanical properties of Thiel-embalmed cadavers better and compare different models of cardiopulmonary resuscitation (CPR). Future research may use this data to translate our findings from bench to bedside. Our recordings may become useful in developing respiratory monitors for CPR, especially in prototyping and testing algorithms of such devices.

Specifications Table

Subject	Emergency Medicine
Specific subject area	Mechanical ventilation under ongoing chest compressions (asynchronous CPR)
Type of data	Table Software Figure
How data were acquired	Data were recorded by a mass flow meter (SFM3000, Sensirion AG, Staefa, Switzerland) and differential pressure sensors (DLVR-L60D, All Sensors Corporation, Morgan Hill, California) each was connected to a Raspberry Pi (Raspberry Pi 3 B+, Raspberry Pi Foundation, Cambridge, United Kingdom), respectively. All Raspberry Pis were integrated into a local network-time-protocol cluster ensuring synchronization and compensation for clock drift. Chest compressions were performed and recorded using an automated chest compression device (Corpuls CPR, GS Elektromedizinische Geräte G. Stemple GmbH, Kaufering, Germany). Data processing was automated by a script in MATLAB (MathWorks, Natick, Massachusetts, United States).
Data format	Raw Processed Filtered
Parameters for data collection	Time stamp of sensor recording: unix time [milliseconds]; flow: standard litre per minute (at 20°C and 1013 mbar), pressure: cmH20 (as differential presseure refrenced to ambient pressure); time stamps from laboratory book: UTC+2 [hh:mm:ss.msec], compression depth [millimetre], compression force [newton]
Description of data collection	Data were obtained from six human cadavers embalmed by the Thiel's method [2,3]. Cadavers were used as a model of asynchronous CPR for a crossover study of three different ventilators. Measurements and interventions have been conducted following an exact study protocol [1].
Data source location	Institution: Division of Macroscopic and Clinical Anatomy, Medical University of Graz City: Graz Country: Austria Latitude and longitude: 47.07778678, 15.44756413 Metres above sea level: 371 m
Data accessibility	Repository name: Data for: Do emergency ventilators deliver preset tidal volumes? - Mechanical ventilation in a human cadaver model of asynchronous cardiopulmonary resuscitation.
	Data identification number: 10.17632/vh4tdsscns.1 Direct URL to data: https://data.mendeley.com/datasets/vh4tdsscns/1 Repository name: ThielViewer - an interactive visualization tool for: Reliability of mechanical ventilation during continuous chest compressions: a crossover study of transport ventilators in a human cadaver model of CPR. Data identification number: 10.17632/43h7zzp67k.3 Direct URL to data: https://data.mendeley.com/datasets/43h7zzp67k/3
Related research article	S. Orlob, J. Wittig, C. Hobisch, D. Auinger, G. Honnef, T. Fellinger, R. Ristl, O. Schindler, P. Metnitz, G. Feigl, G. Prause, Reliability of mechanical ventilation during continuous chest compressions: a crossover study of transport ventilators in a human cadaver model of CPR, Scand J Trauma Resusc Emerg Medicine. 29 (2021) 102. https://doi.org/10.1186/s13049-021-00921-2.

Value of the Data

• •Cardiac arrest is one of the leading causes of death in humans. Mechanics of resuscitation measures are poorly understood. This article provides highly detailed biomechanical data of ventilations and chest compressions in a human cadaver model of CPR.
• •Researchers can use this data to better understand interactions between ventilations and chest compressions. Engineers can use this data to develop respiratory monitors for CPR.
• •This data can be used to understand mechanics of CPR better and maximize comparability of results from different studies.
• •Insights into CPR mechanics may lead to improved resuscitation measures and higher survival rates in cardiac arrest.

1. Data Description

The data we report is sorted in a hierarchical folder structure in a Mendeley Data repository [4]. On top level data are separated into raw and processed data. On second level raw data are sorted into six folders, for each cadaver respectively.

Each folder contains five CSV-files with timeseries data from the used sensors:

• •flow
• •airway pressure
• •esophageal pressure
• •depth of chest compression piston
• •force of chest compression piston

Data in the CSV-files are listed in two columns. The first column lists timestamps in milliseconds unix time and the second column the corresponding sensor reading. While recording systems of flow, airway pressure, and esophageal pressure were synced in a network-time-protocol cluster; both signals extracted from Corpuls CPR (force and depth) are affected with a time offset and a clock drift. Units of measurements are given within a header row in the CSV-file.

Additionally, each folder contains an Excel-file with parameters and timestamps from the laboratory notebook. The coloration is in accordance with the study flowchart from the research article [1].

Derived datasets are stored in the folder “processed data”.

The file “complete_dataset.csv” contains the dataset of volumes and calculated respiratory metrics reported in the original research article [1]. Each row provides data on a single respiratory cycle during CPR. In addition to the three crossover periods, with six sequences of CPR per cadaver, this includes an initial sequence of simulated CPR to assess mechanical properties (sequence = 0). Columns in this dataset and the used units of measurement are described in detail in Table 1.

Table 1.

Columns in the derived dataset with detailed explanations.

Column Number	Variable Name	Units of measurement	Value
1	ID		ID of cadaver
2	height	cm	height of cadaver
3	sex		sex of cadaver
4	weight	kg	predicted body weight
5	device		model of ventilator
6	device no.		position in chronological order of the ventilator in crossover study; corresponds to number of crossover period
7	timestamp	dd-MMM-yyyy hh:mm:ss.fff	time stamp of the peak net tidal volume
8	breath no.		numerical order of ventilation within sequence
9	sequence		chronological order of two-minute-long sequence of simulated CPR
10	Vt_opt [mL]	mL	preset tidal volume
11	Ventilatory rate preset [min−1]	min−1	preset ventilatory rate
12	I:E preset		preset inspiratory to expiratory ratio
13	Pmax preset [cmH₂O]	cmH₂O	preset maximal airway pressure
14	PEEP preset [cmH₂O]	cmH₂O	preset positive endexpiratory pressure
15	Vt_insp [mL]	mL	net inspiratory tidal volume
16	Vt_exp [mL]	mL	net expiratory volume
17	V_insp,sum [mL]	mL	cumulative inspiratory tidal volume
18	V_insp,missed [%]	%	standardized deviation of net inspiratory tidal volume from preset tidal volume
19	V_exp,sum [mL]	mL	cumulative expiratory tidal volume
20	V_exp,missed [%]	%	standardized deviation of net expiratory tidal volume from preset tidal volume
21	pAw, peak [cmH₂O]	cmH₂O	peak airway pressure
22	pOes, peak [cmH₂O]	cmH₂O	peak esophageal pressure
23	flow peak [L/min]	L/min	maximal inspiratory flow
24	Incomplete chest compression		dichotomous variable indicating incomplete chest compression
25	C_dyn [ml/cmH₂O]	mL/cmH₂O	dynamic compliance
26	p_trans [cmH₂O]	cmH₂O	peak transpulmonary pressure
27	p Aw peak no Compression [cmH₂O]	cmH₂O	peak airway pressure during decompression phase
28-32	Insp comp X [mL]	mL	inspiratory reversed airflow
33	Insp comp sum [mL]	mL	cumulative inspiratory reversed airflow
34-48	Exp comp X [mL]	mL	expiratory reversed airflow
49	Exp comp sum [mL]	mL	cumulative expiratory reversed airflow

Concomitantly, the folder contains an exclusion list as CSV-file. The first two columns – “ID” and “timestamp” - serve as unique identifiers of a ventilatory cycle, in accordance with the above described dataset. The third column - “inclusion (Y/N)” - documents consent of the two investigators (J.W. & S.O.), whether CPR was carried out in accordance with the study protocol. Hence, indicating whether the ventilatory cycle had to be excluded from further analysis.

Furthermore, we provide a software tool to review continuous ventilatory tracings from raw data in detail, named “ThielViewer” [5]. This plotting tool is compiled for Windows operating systems and uses MATLAB Runtime 9.5, R2018b (MathWorks, Natick, Massachusetts, United States). For simplified installation, we also provide a web installer within the repository to deploy the necessary runtime.

Once executed, raw data from the data repository [4] can be plotted via the menu “Load data” > “select case”. Within the appearing dialog window, the folder containing the raw data, as CSV-files of a cadaver must be selected. The tool handles data according to the file nomenclature and generates a MATLAB datafile in the selected folder, which allows accelerated replotting of the data.

The tool plots data in a plot of three subplots. The top subplot shows airway and esophageal pressure, plus the derived transpulmonary pressure. The middle subplot shows flow and transpulmonary pressure. The bottom subplot shows tracings of volumes derived from flow. These are net volume, cumulative inspiratory volume, and cumulative expiratory volume. All volumes are reset to zero with beginning of a new ventilatory cycle.

The menu “Options” allows to highlight phases of the measurements by time stamps from the laboratory notebook (s. Fig. 1). The plots are fully interactive, can be rescaled, zoomed in, and panned back and forth. The derived inspiratory net tidal volume is indicted by a red circle with the corresponding “breath no.” (s. Table 1).

Fig. 1.

Window of the plotting tool. With highlights of experiment phases activated (left) and deactivated (right).

Supplementary Data Files

The interactive plots are published alongside this article as supplementary material, for each cadaver respectively. These tracings are stored in MATLAB figure file format. Therefore, they can be visualized with MATLAB (MathWorks, Natick, Massachusetts, United States).

2. Experimental Design, Materials and Methods

An experimental study in embalmed cadavers with three transport ventilators was carried out to assess their reliability in delivering preset tidal volumes during active chest compressions.

Three ventilators MEDUMAT Standard² (WEINMANN Emergency Medical Technology GmbH + Co. KG, Hamburg, Germany), Oxylog 3000 plus (Drägerwerk AG & Co. KGaA, Lübeck, Germany) and Monnal T60 (Air Liquide Medical Systems, Antony Cedex, France) providing volume controlled continuous mandatory ventilation (VC-CMV) were investigated in a crossover study. The study protocol describing the preparations, interventions, and measurements can be found in the supplementary material of the research article [1]. The recording was carried out throughout the experiment, including preparations.

2.1. Cadavers & instrumentation

Cadavers embalmed by a method invented by Walter Thiel [2,3] were selected at random from the storage basins and positioned supine at ambient room temperature of 23°C.

Thiel cadavers have been previously established as a model to study respiratory mechanics during CPR [6]. In comparison to other cadaver models, the Thiel embalmed cadavers were found to have airway resistance, chest wall and lung compliances comparable to critical ill patients [7].

The cadavers were endotracheally intubated. A mass flow meter (SFM3000, Sensirion AG, Staefa, Switzerland) was placed distal to a heat and moisture exchange filter, which was connected to the endotracheal tube. At the side port of the filter, a differential pressure sensor (DLVR-L60D, All Sensors Corporation, Morgan Hill, California) was installed. An oesophagal balloon catheter was inserted and connected to another differential pressure sensor. Both differential pressure sensors were zeroed to atmospheric pressure.

Raw signals were recorded with separate small single-board computers (Raspberry Pi 3 B+, Raspberry Pi Foundation, Cambridge, United Kingdom) which were connected to the sensors. All computers were arranged in a network-time-protocol cluster and synced with each other.

2.2. Interventions

An intensive care ventilator (HAMILTON-C6, Hamilton Medical Inc., Bonaduz, Switzerland) was used to perform initial recruitment of the lung. Two quasi-static inflation manoeuvres were performed using a top pressure of 25 and 30 cmH₂O, followed by a 15-minute mechanical ventilation interval. Repeated pressure-volume curves (P/V loop) were obtained to monitor respiratory mechanics.

The assessment of thoracic and lung characteristics was conducted during three separate two-minute-long sequences.

• 1.only ventilations were provided
• 2.only chest compressions were provided
• 3.a bundle of chest compressions, and ventilations was provided

6 mL/kg predicted body weight was used as the target tidal volume. The closest possible value was used as the preset tidal volume for the respective ventilator (Vt_set). An automated chest compression device (Corpuls CPR, GS Elektromedizinische Geräte G. Stemple GmbH, Kaufering, Germany) provided chest compression with a frequency of 103 min⁻¹, and a compression depth of 5 cm. Data were extracted from the internal recording and exported as CSV timeseries files.

The endotracheal tube was clamped whenever a ventilator was disconnected.

2.3. Data processing

Calculations were done according to formulas in Table 2.

Table 2.

Used calculations to derive ventilatory metrics from raw data.

Variable Name	Performed Calculations to Derive Values from Input	Additional Explanation
weight	● female $height-100cm-10\%$ ● male $height-100cm-5\%$
device no.		[0 - 3]; 0 represents initial sequence during measurement of mechanical properties
sequence		[0 – 6]; 0 represents initial sequence during measurement of mechanical properties; each crossover period consists of two sequences
Vt_opt [mL]	$V{t}_{opt}=6mL/kg*weight$	adjusted to the closest settings available at the specific ventilator
Vt_insp [mL]	$V{t}_{insp}=\underset{t_0}{\overset{t_{max}}{\int }}flow\left(t\right)dt$	integral of flow from the start of inspiration ($t_0$) to time of maximal net tidal volume ($t_{max}$)
Vt_exp [mL]	$V{t}_{exp}=\underset{t_{max}}{\overset{t_{end}}{\int }}flow\left(t\right)dt$	integral of flow from time of maximal net tidal volume ($t_{max}$) to the start of the next inspiration phase ($t_{end}$)
V_insp,sum [mL]	$flo{w}_{pos}\left(t\right)=\{\begin{array}{c}flo{w}_{pos}\left(t\right)=flow\left(t\right)\text{for}flow\left(t\right)>0\\ flo{w}_{pos}\left(t\right)=0\text{for}flow\left(t\right)\le 0\end{array}$ $V_{insp,sum}=\underset{t_0}{\overset{t_{end}}{\int }}flo{w}_{pos}\left(t\right)dt$	Integral of the positive (inspiratory) flow from the start of inspiration ($t_0$) to the begin of next inspiration phase ($t_{end}$)
V_insp,missed [%]	$V_{insp,missed}=\left(\frac{V{t}_{insp}-V{t}_{opt}}{V{t}_{opt}}\right)*100$	percentage deviation of inspiratory net tidal volume from Vtopt
V_exp,sum [mL]	$flo{w}_{neg}\left(t\right)=\{\begin{array}{c}flo{w}_{neg}\left(t\right)=flow\left(t\right)\text{for}flow\left(t\right)<0\\ flo{w}_{neg}\left(t\right)=0\text{for}flow\left(t\right)\ge 0\end{array}$ $V_{exp,sum}=\underset{t_0}{\overset{t_{end}}{\int }}flo{w}_{neg}\left(t\right)dt$	Integral of the negative (expiratory) flow from the begin of inspiration ($t_0$) to the begin of next inspiration phase ($t_{end}$)
V_exp,missed [%]	$V_{exsp,missed}=\left(\frac{V{t}_{exp}-V{t}_{opt}}{V{t}_{opt}}\right)*100$	percentage deviation of expiratory net tidal volume from Vtopt
pAw, peak [cmH₂O]	maxima of airway pressure
pOes, peak [cmH₂O]	maxima of esophageal pressure
flow peak [L/min]	maxima of flow
Incomplete chest compression	positive when one of the following criteria of esophageal pressure is fulfilled:  ● difference of maxima between two strokes (chest compression) > 20cmH2O  ● interval between two maxima > 0.6s  ● less than 8 maxima during ventilation (< 8 chest compressions)	Y: incomplete chest compressions by mentioned criteria; N: no indication for incomplete chest compressions
C_dyn [ml/cmH₂O]	$c_{dyn}=\frac{V{t}_{insp}}{p_{Aw,peak}}$
p_trans [cmH₂O]	maxima of $p_{trans}\left(t\right)=p_{Aw}\left(t\right)-p_{Oes}\left(t\right)$
p Aw peak no Compression [cmH₂O]	maxima of airway pressure during decompression phase
Insp comp X [mL]	$V_{insp,comp,X}=\left|{\int }_{t_{0,X}}^{t_{1,X}}flo{w}_{neg}\left(t\right)dt\right|$ Indexed by the chronological number of chest compression during inspiration	integral of negative air flow during inspiration and chest compression
Insp comp sum [mL]	$V_{insp,comp,sum}={\sum }_{k=1}^n{V}_{insp,comp,k}$ with n as number of chest compressions during inspiration of one breath
Exp comp X [mL]	$V_{insp,comp,X}=\left|{\int }_{t_{0,X}}^{t_{1,X}}flo{w}_{pos}\left(t\right)dt\right|$ Indexed by the chronological number of chest compression during expiration	integral of positive air flow during expiration and chest compression
Exp comp sum [mL]	$V_{exp,comp,sum}={\sum }_{k=1}^n{V}_{exp,comp,k}$ with n as number of chest compressions during expiration of one breath

Ethics Statement

All bodies were donated to the Chair of Macroscopic and Clinical Anatomy of the Medical University of Graz, under the strict rules of the anatomical donation program according to the Styrian burial law for scientific purpose. Hence, no additional approval by the local ethical board was necessary.

Funding Source

This study was funded by the Austrian Association of Emergency and Disaster Medicine (abbr.: ÖNK) with the “Reinhard Malzer Award”.

Declaration of Competing Interest

SO has received the “Reinhard Malzer Award” as funding for this study. But the association did not interfere with any steps towards this article.

GP has given a talk at a national symposium, invited by RWM Medizintechnik GmbH.

All other authors have no personal conflict of interest.

Medical devices and equipment used in this study were kindly lent by the following companies: CHEMOMEDICA Medizintechnik und Arzneimittel VertriebsgmbH, Löwenstein Medical Austria GmbH, Sanitas GmbH, GS Elektromedizinische Geräte G. Stemple GmbH, Dräger Austria GmbH, WEINMANN Emergency Medical Technology, RWM Medizintechnik GmbH. No company or manufacturer had influence on the study protocol, statistical analyses, nor were involved in writing of this paper. Dr. Orlob reports grants from Österreichische Gesellschaft für Notfall- und Katastrophenmedizin (abbr.: ÖNK), during the conduct of the study.

Dr. Prause reports a talk at a national symposium, invited by RWM Medizintechnik GmbH.

All other authors have no personal conflict of interest.

Data Availability

• Data for: Do emergency ventilators deliver preset tidal volumes? - Mechanical ventilation in a human cadaver model of asynchronous cardiopulmonary resuscitation (Original data) (Mendeley Data).

• ThielViewer - an interactive visualization tool for: Do emergency ventilators deliver preset tidal volumes? - Mechanical ventilation in a human cadaver model of asynchronous cardiopulmonary resuscitat (Original data) (Mendeley Data).