Anesth Analg 2008; 107:541-544
© 2008 International Anesthesia Research Society
doi: 10.1213/ane.0b013e31817e6778
TECHNOLOGY, COMPUTING, AND SIMULATION
Section Editor:
Jeffrey M. Feldman
The Performance of Six Pulse Oximeters in the Environment of Neuronavigation
Alexander M. Mathes, MD,
Sascha Kreuer, MD,
Sven O. Schneider, MD,
Stephan Ziegeler, MD, and
Ulrich Grundmann, MD
From the Department of Anesthesiology, Critical Care and Pain Medicine, Saarland University Hospital, Homburg (Saar), Germany.
Address correspondence and reprint requests to Ulrich Grundmann, MD, Saarland University Hospital, Department of Anesthesiology, Critical Care and Pain Management, Kirrberger Straβe, D-66421 Homburg (Saar), Germany. Address e-mail to ulrich.grundmann{at}uks.eu.
 |
Abstract
|
|---|
BACKGROUND: Although the use of pulse oximeters may be regarded a standard of care for monitoring anesthesia procedures, these monitors may be susceptible to various kinds of disturbances. Recently, it was suggested that neuronavigation equipment may interfere with pulse oximeter accuracy. In this study, we evaluated the effect of a neurosurgical image guidance system on the performance of six different pulse oximeters. Two simple shielding methods were evaluated.
METHODS: Twenty healthy, adult, nonsmoking volunteers were equipped with six different pulse oximeters on both hands. Baseline values for heart rate, arterial oxygen saturation, and signal quality were assessed. After activation of the Brain Lab VectorVision Neuronavigation System, the effects on signal quality and saturation recognition were evaluated. Measurements were repeated using two different shielding techniques, a cotton blanket and aluminum sheets.
RESULTS: Activation of the image guidance system resulted in a significant disturbance of signal quality and saturation detection, which was partially reversible by both shielding techniques. Significant differences were noted among the six brands of pulse oximeters for signal quality (P < 0.001) and saturation recognition (P < 0.001), and for the response to shielding methods (P < 0.001). Coverage of the probes with aluminum foil resulted an in undisturbed saturation recognition in all subjects with almost all monitors.
CONCLUSIONS: Infrared pulse waves from neurosurgical navigation equipment may interfere with pulse oximeter measurements. Shielding the probe with aluminum foil sufficiently eliminated the infrared interference.
|
Introduction
|
|---|
Pulse oximetry has evolved into a standard technique for monitoring arterial oxygen saturation (Spo2) in patients undergoing anesthesia and monitored sedation.1,2 Pulse oximeters are easy to use, do not require calibration and provide almost immediate and continuous information regarding changes in Spo2. However, a variety of factors may interfere with the accuracy of measurements or may result in erroneous readings.3 Subject motion,4 skin pigmentation,5 dyshemoglobineamia,6 intravascular dyes,7 and poor perfusion8 have been shown to limit the use of this monitoring technique. Recently, an additional source of interference was reported in a patient who underwent neurosurgery: the use of a frameless stereotactic neurosurgical positioning system (Stealth Station Image Guidance System; Medtronic Sofamor Danek, Memphis) was demonstrated to reduce the accuracy of pulse oximeter monitoring.9 This phenomenon has also been observed in our neurosurgical operating theater during neuronavigation (NN) with a Brain Lab VectorVision Neuronavigation System (Brain Lab, Moorestone, NJ). However, the degree of interference by pulsed infrared waves from neurosurgical image guidance systems may depend on the filtering techniques of the pulse oximeter monitor implied by each manufacturer. Therefore, this study was designed to investigate the effects of NN on the performance of six different brands of pulse oximeters. Further, two simple shielding methods were assessed.
|
METHODS
|
|---|
Twenty adult, healthy, nonsmoking men and women volunteers participated in this study, with informed consent and after IRB approval. Each subject was monitored with six pulse oximeter finger probes that were mounted in a random manner on the index, middle, and ring fingers of both hands, while subjects were breathing room air. The six different pulse oximeters tested are listed in Table 1.
All experiments were conducted in the same neurosurgical operating theater with constant ambient light and temperature. The subjects participated one at a time, placing the fingers on predetermined marks on the operating table. The oximeter sensors were applied according to the instructions of the manufacturer and connected to their respective instruments. Interference among the probes was prevented by using plastic shielding between each finger.
For each subject and monitor, heart rate (HR), Spo2 and signal quality (SQ) was determined in the course of the experiment. All six monitors were evaluated by three observers at the same time, each observer responsible for two monitors. SQ was defined as follows: No signal detection was defined as no pulse wave signal present during the whole time of the assessment. Severe noise was defined as a disruption of the signal for one or more episodes of more than 15 s, whereas moderate artifacts was defined as one or more episodes of <15 s of signal interference. Normal readings was defined as an undisrupted signal for the whole time of the assessment. Undisturbed saturation recognition was defined as a saturation reading present for the whole time of the assessment.
Baseline values were recorded for 5 min. The cameras of a neurosurgical stereotactic guidance system (Brain Lab VectorVision Neuronavigation System; Brain Lab, Moorestone, NJ) were then brought into a position at 100 cm above both hands with an unobstructed view of the fingers. The NN system was activated, and for 5 min, the readings for HR, Spo2 and SQ were assessed. Each finger was first wrapped with a single layer of a cotton blanket (thickness: 3.0 mm), and after removal of the blankets, they were wrapped with a single commercial grade sheet of aluminum foil (thickness: 0.01 mm). Measurements for HR, Spo2 and SQ were obtained for 5 min of coverage, respectively, and were repeated for 5 min after deactivation of the NN system, after the shielding materials had been removed.
All data are presented either as mean ± sd (SD), or as percentage of subjects. Statistical analysis was performed using the software SigmaPlot 9.0 (Systat, Erkrath, Germany). Differences between incidences (percentages) were evaluated applying a  2-analysis. Spo2 data were compared using one-way analysis of variance (ANOVA). A P value of < 0.05 was considered significant.
|
RESULTS
|
|---|
Signal Quality
Data for SQ analyses are given in Table 2. All monitors displayed normal pulse oximeter signals at the beginning of the experiment. After activation of the NN equipment, all monitors were at least partially affected by signal interference. Percentages of subjects with undisturbed signal recognition for each monitor are shown in Figure 1. Interference with pulse oximeter SQ was different among manufacturers (2-analysis; P < 0.001 for NN with no shielding, P < 0.001 for NN with blanket coverage, and P < 0.001 for NN with aluminum sheet). Shielding with a cotton blanket resulted in an improvement of SQ for all monitors except the Dash pulse oximeter (2-analysis; P < 0.01 versus unshielded readings). Signal disturbances were further diminished by coverage with aluminum foil for the Dash, In vivo, Dräger and Philips pulse oximeters (P < 0.01 versus blanket coverage).
View larger version (27K):
[in this window]
[in a new window]
|
Figure 1. Percentages of subjects with undisturbed signal recognition. Performance of pulse oximeters was different among manufacturers, irrespective of the method of shielding during neuronavigation (NN) (2 analysis; P < 0.001). A pound sign (#) indicates P < 0.01 versus NN without coverage; an asterisk (*) indicates P < 0.01 versus blanket coverage.
|
|
Arterial Oxygen Saturation Reading
Spo2 readings displayed normal values for all subjects at the beginning of the experiment (Spo2 98.7% ± 1.3 sd). No significant differences were observed among the monitors for baseline Spo2 readings (one-way ANOVA; P = 0.99). After activation of the neurosurgical imaging guidance system, 5 of 6 monitors were unable to display Spo2 in some of the subjects without shielding (Fig. 2); only the Dräger monitor was able to measure Spo2 in all subjects, irrespective of NN equipment in use. Under the influence of NN, the rate of undisturbed Spo2 detection was different among the pulse oximeter monitors used in this study (P < 0.001 for NN with no shielding, P < 0.001 for NN with blanket coverage, and P < 0.001 for NN with aluminum foil). Shielding with a single layer of a cotton blanket improved saturation detection significantly for the Dash, In vivo, Tuffsat, BCI and Philips monitors (P < 0.01 versus no shielding). Coverage with aluminum foil further attenuated interference from NN in the Dash, In vivo and Tuffsat monitor; however, this was only significant for the Dash monitor (P < 0.001).
View larger version (33K):
[in this window]
[in a new window]
|
Figure 2. Percentages of subjects with undisturbed saturation recognition. Performance of pulse oximeters was different among manufacturers (2 analysis; P < 0.001). A pound sign (#) indicates P < 0.01 versus NN without coverage; an asterisk (*) indicates P < 0.01 versus blanket coverage.
|
|
Whenever a Spo2 reading was present, no statistical difference was noted between baseline values and the ones obtained during NN (one-way ANOVA; P = 0.99). In two subjects, Spo2 readings showed a noticeable decrease during NN in a single monitor only, irrespective of a shielding technique in use (Subject 13; In vivo monitor: Spo2 100% without NN versus 84% with NN. Subject 19; Dash monitor: Spo2 99% without NN versus 78% with NN).
|
DISCUSSION
|
|---|
This study demonstrates that the use of a stereotactical surgical image guidance system may interfere with pulse oximeter monitoring. Activation of a neurosurgical positioning system disturbed pulse oximeter SQ in all monitors, but not in all subjects investigated. Further, interference of the NN equipment with Spo2 readings was detected in almost all monitors in this study. In two subjects, activation of the NN system resulted in erroneous pathological readings of Spo2. Simple shielding techniques may be used to attenuate these effects; coverage with a single aluminum sheet seems to offer a better protection than a single layer of a cotton blanket.
Infrared light pulses from the Brain Lab Vector Vision Neuronavigation System appear to interfere with the detection of infrared light signals emitted from pulse oximeter probes. The wavelengths used for measurement of the extinction of oxygenated hemoglobin (HbO2) are typically in the range of 910–940 nm, depending on the pulse oximetry monitor used, and show an overlap with the infrared pulse waves of NN equipment, emitting a spectrum from 775 to 1000 nm (Fig. 3) [personal communication with Brain Lab Germany and NBI Europe]. As a consequence, the calculation of the absorption ratio between hemoglobin and HbO2 may be at least partially disrupted. This observation is in line with previous reports of the disturbance of pulse oximetry by infrared light waves, e.g., from infrared heat lamps.10 In our study, signal waveform detection was disturbed in 20%–100% of the subjects, depending on the monitor investigated. Shielding with aluminum foil resulted in an improvement of the waveform signal, allowing for more than 80% of undisturbed signals in all monitors, except the Dash pulse oximeter (15%). These findings confirm the observations made by van Oostrom et al., reporting similar disturbances with a different model of NN equipment (Stealth Station Image Guidance System, Medtronic Softamor Danek, Memphis).9
View larger version (27K):
[in this window]
[in a new window]
|
Figure 3. The infrared pulse wave spectrum of the Vector VisionTM neuronavigation camera interferes with the measurement of oxygenated hemoglobin (HbO2), typically performed at 910–940 nm (B), but not with the measurement of hemoglobin (Hb), usually measured at 660 nm (A).
|
|
Spo2 recognition was at least partially affected by the surgical guidance image system in almost all pulse oximeters; only the Dräger monitor was not disturbed by NN in this respect. Shielding with a single layer of blanket or aluminum foil resulted in 85%–100% of undisturbed saturation readings. With the exception of the Dash monitor, coverage with a single layer of a cotton blanket was sufficient to allow for 95% of unaffected saturation readings in all monitors in this study.
Interestingly, the interference of NN with SQ does not necessarily result in a disruption of saturation recognition. Although the Dräger and Philips monitors display severe disturbances regarding SQ, these pulse oximeters yield the best results concerning undisturbed saturation recognition. This finding may be attributable to specific filtering algorithms of each monitor. However, the technical details of these filtering techniques are not released by the manufacturers, making a comparison impossible.
An obvious variation of Spo2 from baseline values was noticed in two single cases during NN, with different monitors in use. Thus, the decrease in saturation readings to 80% after activation of the Stealth System, as reported by van Oostrom et al.,9 may be present in selected subjects. Although most saturation readings were unaffected by the stereotactical guidance system in our study, it should be recognized that an apparent decrease in Spo2 may result in serious changes inpatient care. Therefore, this effect needs to be kept in mind when sudden changes in Spo2 are displayed by a pulse oximeter in the environment of NN.
Not all measurements within a specific monitor were equally affected by NN among subjects. Since the ambient conditions during this study were controlled and interference among sensors was prevented, it may be that the varying interference within one monitor was caused by infrared pulse wave reflections from the subject, due to differences in body configuration. Consequently, it may be assumed that small changes in the environment of NN, e.g., movement of personnel or equipment, may result in possible sources of pulse oximeter disturbance.
We conclude that the use of image guidance systems for NN may potentially interfere with pulse oximeter monitoring. The use of a single layer of aluminum foil seems to offer a reliable protection from this effect. Different pulse oximeter monitors show a variation in susceptibility to the disturbance caused by NN equipment. Interindividual differences may influence the extent of interference, making it difficult to predict disturbances in a specific setting.
|
ACKNOWLEDGMENTS
|
|---|
We would like to thank Stefan Heinersdorff from Brain Lab Germany, and NBI Europe, for the technical details on neuronavigation.
|
Footnotes
|
|---|
Supported by Departmental funds.
Accepted for publication April 15, 2008.
|
REFERENCES
|
|---|
- ASA House of Delegates. Standards for basic anesthetic monitoring. In: ASA Standards, Guidelines and Statements, October 2005. Available online: www.asahq.org/publicationsandservices/sgstoc. htm, last accessed on November 6, 2007
- Knape JT, Adriaensen H, van Aken H, Blunnie WP, Carsson C, Dupont M, Pasch T. Board of Anaesthesiology of the European Union of Medical Specialists. Guidelines for sedation and/or analgesia by non-anaesthesiology doctors. Eur J Anaesthesiol 2007;24:563–7[Web of Science][Medline]
- Jubran A. Pulse Oximetry. Crit Care 1999;3:R11–R17[Web of Science][Medline]
- Barker SJ, Shah NK. Effects of motion on the performance of pulse oximeters in volunteers. Anesthesiology 1996;85:774–81[Web of Science][Medline]
- Bickler PE, Feiner JR, Severinghaus JW. Effects of skin pigmentation on pulse oximeter accuracy at low saturation. Anesthesiology 2005;102:715–9[Web of Science][Medline]
- Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia on pulse oximetry and mixed venous oximetry. Anesthesiology 1989;70:112–7[Web of Science][Medline]
- Scheller MS, Unger RJ, Kelner MJ. Effects of intravenously administered dyes on pulse oximetry readings. Anesthesiology 1986;65:550–2[Web of Science][Medline]
- Clayton DG, Webb RK, Ralston AC, Duthie D, Runciman WB. A comparison of the performance of 20 pulse oximeters under conditions of poor perfusion. Anaesthesia 1991;46:3–10[Web of Science][Medline]
- van Oostrom JH, Mahla ME, Gravestein D. The Stealth Station Image Guidance System may interfere with pulse oximetry. Can J Anaesth 2005;52:379–82[Web of Science][Medline]
- Brooks TD, Paulus DA, Winkle WE. Infrared heat lamps interfere with pulse oximeters. Anesthesiology 1984;61:630–1[Web of Science][Medline]
|
Top
Abstract
Introduction
