Test Report PM V205062021
Page 2 of 5
Device
The device tested was the OB1203SD-RL-EVK system. This is a spot check
device that is not mechanically suited to continuous use. For this test it was fitted
with a rubber boot to ensure that the finger remained positioned during the test.
Methods – UCSF Hypoxia test
The group of volunteers for the study consisted of 12 subjects 8 males and 4
females.
A radial arterial cannula was placed in either the left or right wrist of each subject.
Blood gas analysis to determine oxyhemoglobin saturation was performed on an
ABL90 FLEX multi-wavelength oximeter (Hemoximeter, Radiometer,
Copenhagen). No subject was anemic (Hemoglobin ≤ 10gm•dl-1).
Each subject had control data taken at the beginning of each experiment, with
two control blood samples drawn while breathing room air. Hypoxia was induced
to different levels of oxyhemoglobin saturation (between 70-100%) by having
subjects breathe mixtures of nitrogen, room air, and carbon dioxide.
Oxyhemoglobin saturation was reduced to a series of targets and stabilized at the
plateau value. Each plateau level of oxyhemoglobin saturation was maintained
for at least 30 seconds. Two arterial blood samples were then obtained,
approximately 30 seconds apart. A total of 24 samples were obtained per subject.
Data were recorded by UCSF Hypoxia Lab and provided for analysis.
Data Set – Collected by Physio Monitor
Data analysis was performed using Microsoft Excel. Files containing the sat
value outputs every second from the test instruments were imported. Co-
oximeter values were imported from the ABL90. Sample timing was entered
manually into the computer used to time & control the test.
In this type of study, the central saturation is changed very quickly. Blood takes
time to circulate in the human body, so the values change at different times for
the different measurement locations. Differences of as much as 20 seconds are
common. In order to minimize the impact of this effect on the data the plateau
times are identified from the test device data and matched to the corresponding
blood samples.
If at any plateau the ABL90 readings differ by more than 2% the readings from
that plateau are discarded. In this study about 10% of the readings were
discarded for this reason. In addition, there were difficulties in identifying plateaus
in the test device data. For this reason, another 61% of the points are missing.
There are 287 remaining data points some of which are from each of the 10
subjects. This is enough to meet the ISO / FDA minimum of 10 subjects and 200
data points.