70
J. Kumar V. and K. A. Reddy
carboxyhaemoglobin, which the oximeter detects as oxyhaemoglobin and thus
overestimates the true concentration of oxyhaemoglobin. Other dyshaemoglobins
such as methaemoglobin and acraboxyhaemoglobin also interfere [52, 53]. Only
Co-Oximetry should be employed if the presence of abnormal haemoglobin is
expected.
Other parameters such as dyes/pigments also affect pulse oximetry measurements.
Methylene blue, which is recommended as a therapy for methemoglobinemia, absorbs
light at 660 nm, like the absorption of Hb. If methylene blue is present in blood, then
the readings obtained on a pulse oximeter will be erroneous [54]. Similar problems can occur with other intravenous dyes such as fluorescein, indocyanine green,
and indigo carmine that are used for therapeutic or diagnostic purposes [55]. Since
pulse oximeters estimate SpO 2 level using red and IR PPGs, clear noise and artefact
free red and IR PPGs are essential. To obtain proper red and IR PPG signals an
adequate arterial pulsation is required. Shock, severe hypotension, vasoconstriction
or hypothermia can lead to significant decrease in peripheral vascular pulsation and
thus can produce unreliable oximeter readings [56]. Another variable that affects
pulse oximetry readings can be due to nail polish on a finger or toe probe. It is
ascertained that oximeter readings of oxygen saturation are affected by black, blue,
and green nail polish. Hence it is better to remove fingernail polish or artificial nails
before placing the pulse oximetry probe. Another technique by which the effect due
to nail polish could be eliminated is by placing the finger probe side to side instead
of top to bottom. For error-free SpO 2 estimation in a pulse oximeter, clean noise-free
and artefact-free red and IR PPG signals with distinct DC and AC parts are necessary.
Errors in the measured values of SpO 2 can also arise in situations that produce low
signal-to-noise ratio on the detected red and IR PPG signals.
Light from sources such as surgical lamps, infrared, xenon and fluorescent lamps
can interfere with oximetry accuracy [57]. If excessive ambient light interference is
present, then both the red and IR PPG signals will be corrupted. However, a recent
study revealed that moderate ambient light has no significant effect on pulse oximetry
readings [58]. Even had the differences been statistically significant, the magnitude
of the differences was small and thus clinically unimportant. Interference with highintensity ambient light can be easily avoided by placing an opaque material (darkcoloured cloth) to surround the probe. Studies on the effect of skin colour on the pulse
oximetry readings revealed that dark skin colour may influence the performance and
concluded that accuracy was slightly less for subjects with dark skin colour than
those with lighter skin colour [49]. Effects of absorption by platelets and plasma
indicate that these have negligible effects on the SpO 2 measurement [59].
Movement of a patient introduces artefacts in the red and IR PPG signals and thus
will cause inaccurate readings in a pulse oximeter [51]. Since the pulsatile (ac part)
component of a PPG is ≈0.1% of total signal amplitude, even a slightest movement
of the patient will disturb the PPG. Additional random pulsations will be added to
the pulsations due to heartbeat. Such additional random pulsations in a PPG due
to motion artefact will result in inaccurate estimation of SpO 2 . A straightforward
solution is to altogether avoid motion artefacts by securing the probe housing the
sources and detectors rigidly to the skin of the patient at a monitoring site such that
J. Kumar V. and K. A. Reddy
carboxyhaemoglobin, which the oximeter detects as oxyhaemoglobin and thus
overestimates the true concentration of oxyhaemoglobin. Other dyshaemoglobins
such as methaemoglobin and acraboxyhaemoglobin also interfere [52, 53]. Only
Co-Oximetry should be employed if the presence of abnormal haemoglobin is
expected.
Other parameters such as dyes/pigments also affect pulse oximetry measurements.
Methylene blue, which is recommended as a therapy for methemoglobinemia, absorbs
light at 660 nm, like the absorption of Hb. If methylene blue is present in blood, then
the readings obtained on a pulse oximeter will be erroneous [54]. Similar problems can occur with other intravenous dyes such as fluorescein, indocyanine green,
and indigo carmine that are used for therapeutic or diagnostic purposes [55]. Since
pulse oximeters estimate SpO 2 level using red and IR PPGs, clear noise and artefact
free red and IR PPGs are essential. To obtain proper red and IR PPG signals an
adequate arterial pulsation is required. Shock, severe hypotension, vasoconstriction
or hypothermia can lead to significant decrease in peripheral vascular pulsation and
thus can produce unreliable oximeter readings [56]. Another variable that affects
pulse oximetry readings can be due to nail polish on a finger or toe probe. It is
ascertained that oximeter readings of oxygen saturation are affected by black, blue,
and green nail polish. Hence it is better to remove fingernail polish or artificial nails
before placing the pulse oximetry probe. Another technique by which the effect due
to nail polish could be eliminated is by placing the finger probe side to side instead
of top to bottom. For error-free SpO 2 estimation in a pulse oximeter, clean noise-free
and artefact-free red and IR PPG signals with distinct DC and AC parts are necessary.
Errors in the measured values of SpO 2 can also arise in situations that produce low
signal-to-noise ratio on the detected red and IR PPG signals.
Light from sources such as surgical lamps, infrared, xenon and fluorescent lamps
can interfere with oximetry accuracy [57]. If excessive ambient light interference is
present, then both the red and IR PPG signals will be corrupted. However, a recent
study revealed that moderate ambient light has no significant effect on pulse oximetry
readings [58]. Even had the differences been statistically significant, the magnitude
of the differences was small and thus clinically unimportant. Interference with highintensity ambient light can be easily avoided by placing an opaque material (darkcoloured cloth) to surround the probe. Studies on the effect of skin colour on the pulse
oximetry readings revealed that dark skin colour may influence the performance and
concluded that accuracy was slightly less for subjects with dark skin colour than
those with lighter skin colour [49]. Effects of absorption by platelets and plasma
indicate that these have negligible effects on the SpO 2 measurement [59].
Movement of a patient introduces artefacts in the red and IR PPG signals and thus
will cause inaccurate readings in a pulse oximeter [51]. Since the pulsatile (ac part)
component of a PPG is ≈0.1% of total signal amplitude, even a slightest movement
of the patient will disturb the PPG. Additional random pulsations will be added to
the pulsations due to heartbeat. Such additional random pulsations in a PPG due
to motion artefact will result in inaccurate estimation of SpO 2 . A straightforward
solution is to altogether avoid motion artefacts by securing the probe housing the
sources and detectors rigidly to the skin of the patient at a monitoring site such that