Study method The data collection will start 1st of February 2021 and will be finished by
30th of April 2021. Patients will be set up with TCM equipment and an Empatica at the
PACU. They will be observed for 2 hours. If an intervention is started before 2 hours and
not completed, recordings will continue until 5 minutes after the intervention is
completed for a maximum of 1 hour (3 hours maximal observation). The sensor will be
placed on the patient's antebrachium. Patients will be set up with TCM equipment as soon
as possible after arrival at PACU. Every clinical procedure entailing changes in
circulatory and/or respiratory support or medication (appendix 1) will be observed and
noted in a timetable (ap-pendix 2). Perfusion index (PI) and a full DASAIM score (Danish
Association of Anesthesiology and Intensive Care Medicine) will be noted every 15
minutes. Every ABG analysis of PaO2, PaCO2 and lactate conducted during the observational
period will be noted.
Age, sex, BMI (Body Mass Index), ASA-score (American Society of Anesthesiologists
physical sta-tus classification), comorbidity (Charlson Comorbidity Index - CCI),
baseline laboratory data (hemo-globin and creatinine) and name of surgical procedure will
be obtained for each patient. REDCap (Research Electronic Data Capture) will be used as a
data management system for the project. The study is a part of the Wireless Assessment of
Respiratory and circulatory Distress (WARD) project.
Study materials All patients will be monitored with a TCM-5 Flex monitor connected to a
Sensor 84 (Radiometer Medical Aps, Brønshøj, Denmark). The sensor consists of a
Stow-Severinghaus-type TcPCO2 electrode combined with a Clark-type TcPO2 electrode, with
values per second for TcPO2, TcPCO2 in kPa and heat effect measured in units of mW. The
sensor temperature will automatically be heated to 43.5℃ and calibrated. The sensor
membrane will be changed every 28th day. There will be a 20-minute stabilization period
from patient setup to valid data capturing. The use of sensors and monitors will be
according to the manufacturer's instructions.
As a supplement, patients will be wearing an Empatica, as a measurement for the activity
of the autonomic nervous system. The Empatica tracks activity, heartrate, blood volume
pulse, skin temper-ature and skin conductance. The Empatica E4 consist of a PPG sensor
and an EDA sensor. Data from the E4 will be processed in further studies and is not a
part of the statistical analysis in this protocol.
Statistics The primary analysis is changes in TCM readings, and the association between
clinical procedures, tested using paired statistics comparing both average (mean) TcPO2
and TcPCO2 during 5 minutes before an intervention versus 5 minutes after an
intervention. Depending on the data distribution, either t-test or Wilcoxon test will be
used.
Secondary analysis includes:
Changes in TCM and association with specific intervention including grouping into
circulatory and respiratory groups.
Agreement between TcPO2 and PaO2, TcPCO2 and PaCO2. Agreement between the two
tech-niques will be assessed using Bland-Altmann analysis including description of
mean differ-ence and limits of agreement(17). In case of left-skewed data, log
transformation will be ap-plied. A four-quadrant plot will be constructed to assess
trending ability and a concordance rate will be calculated.
Tertiary analysis includes:
An ANCOVA (analyses of co-variance) of the primary and secondary outcomes, with the
following predefined variables: age, sex, BMI, ASA-score, and comorbidity, baseline
labora-tory data (hemoglobin and creatinine).
Agreement between perfusion-index and TCM. Agreement between the two techniques will
be assessed using Bland-Altmann analysis including description of mean difference
and limits of agreement. In case of left-skewed data, log transformation will be
applied. A four-quadrant plot will be constructed to assess trending ability and a
concordance rate will be calculated.
All results will be stated with their 95% confidence interval (CI). The significance
level will be 5%, meaning p-values under 0.05 will be acknowledged as statistically
significant. All analyses will be performed using Python 3.7 and the software program
PyCharm.