Pulse oximetry is ubiquitous in anesthetic practice and is an American Society of Anesthesiologists standard monitor.1 The pulse oximeter probe can be either reusable or disposable. The disposable adhesive oximeter probe is popular because it is easy to use, more resistant to motion artifact, eliminates the possibility of patient-to-patient contamination, is less likely to come off if the patient moves,2 and may be more economical in some settings. However, the reusable probe differs in its usability from the disposable version. Because of its rigid structure and hinged construction, the design of the reusable probe fosters proper alignment of the light-emitting diode (LED) with the photodetector. In contrast, alignment of the LED and photodetector of the disposable probe is entirely in the control of the operator, and depends on how carefully the probe is applied to the patient’s finger. Herein, we report 3 cases in which misaligned disposable adhesive pulse oximeter probes gave false intraoperative saturation readings (Spo2) despite high-quality plethysmographic tracings, which, in turn, led to unnecessary invasive interventions or changes of anesthesia management. In each instance, reapplication of a correctly aligned new probe was the “curative” intervention.
A 39-yr-old woman was scheduled for a left total knee replacement. A disposable adhesive pulse oximeter probe (Masimo, Irvine, CA) was placed on her finger. Her preoperative Spo2 was 99% while breathing 100% oxygen. After routine anesthesia induction, an endotracheal (ET) tube was easily inserted and taped after confirming bilateral breath sounds. During maintenance of anesthesia, the fraction of inspired oxygen (Fio2) was kept at 30%. The patient’s intraoperative pulse oximeter reading gradually declined to 88%, although the plethysmographic tracing quality indicator continued to indicate good signal quality. Breath sounds were again found to be clear and equal bilaterally. The ET tube was suctioned, but only scant secretions were recovered. Bronchoscopy confirmed that the tip of the ET tube was 2.5 cm above the carina, and the entire tracheobronchial tree was clear. The anesthesia team reduced nitrous oxide to allow an increase of the Fio2 to around 90% to keep the pulse oximeter reading 97%–99%. About 35 min later, the anesthesia team checked the pulse oximeter probe and found that it was misplaced, with the light source and the detector misaligned axially along the finger by about 15 mm. The “misalignment distance” is defined as the distance between the actual photodetector position (relative to the LED) and the ideal photodetector position (i.e., exactly opposite the LED) (Fig. 1). A new probe was placed with a misalignment distance of 2 mm. The pulse oximeter reading from this properly placed sensor was 100% with an Fio2 of 95%. Subsequently, the Fio2 was gradually reduced to and maintained at 30% by the resumption of nitrous oxide. The saturation reading was 97% or higher from the correctly applied probe throughout the remainder of the case.
A 37-yr-old woman was scheduled for left video-assisted thoracoscopic surgery to remove a 3-cm left upper lobe lung mass. A disposable adhesive oximeter probe was applied on her finger. Her preoperative pulse oximeter reading was 99% while breathing 100% oxygen. After routine anesthesia induction, her pulse oximeter reading remained around 96%–98% with an Fio2 of 97% during fiberoptic bronchoscopy by the surgeon and 97%–99% with an Fio2 of 99% after a right-sided double-lumen tube was inserted. The Fio2 was maintained at 99% during right-sided 1-lung ventilation. The pulse oximeter reading was initially 99%, but gradually declined to 94%, with a concomitant 3-star signal quality indication and stable arterial blood pressure and heart rate. Bronchoscopy confirmed correct position of the right-sided double-lumen tube and a clean tracheobronchial tree. Before the anesthesia team tried dependent positive end-expiratory pressure or insufflation of the nondependent lung with oxygen, they decided to check the pulse oximeter probe and found that it had a misalignment distance (axially along the finger) of 10 mm. A new probe was placed with a misalignment distance of 1 mm. The properly placed pulse oximeter reading was 100% and remained at 100% while using an Fio2 of 99% during 1-lung ventilation throughout the remainder of the case.
A 71-yr-old man was scheduled for a right common femoral endarterectomy. A disposable adhesive pulse oximeter probe was placed on the patient’s left middle finger. His preoperative Spo2 was 98%–99% with 100% oxygen. After routine anesthesia induction, an ET tube was easily passed and taped. Breath sounds were bilateral and equal. The Fio2 was kept at 30% during anesthesia maintenance. The pulse oximeter initially gave a consistent reading of 96%–97% but gradually decreased to 93% with stable arterial blood pressure and heart rate. The Fio2 was increased to 52%, and the Spo2 initially increased to 95% but again gradually decreased to 93%, with stable arterial blood pressure, heart rate, and temperature. The Fio2 was then further increased to 95%, but the pulse oximeter reading remained at 93%. A search for the cause of hypoxia was initiated while temporizing by increasing the Fio2. The inspiratory pressure did not increase significantly, and the capnographic trace was completely normal. Upon auscultation, the breath sounds were clear and equivalent bilaterally, without crackles or wheezes. Bronchoscopy revealed an entirely clear tracheobronchial tree bilaterally.
Monitor data from the above sequence events are shown in Figures 2 through 4. With Fio2 of 29%, the pulse oximeter reading remained at 91%–92% (Fig. 2), and arterial blood pressure, heart rate, and temperature were all clinically acceptable and stable. Throughout this sequence of events, the Spo2 signal quality indicator registered 3 stars (Fig. 2), indicating a good signal-to- noise ratio. At this point, an arterial blood gas from the right radial arterial catheter showed a Pao2 of 141 mm Hg, consistent with what would be expected from normal gas exchange at the set Fio2. This Pao2 was inconsistent with the measured Spo2 of 91%–92%. The arterial blood gas result raised the suspicion that the pulse oximeter might be providing erroneous data. Therefore, a new disposable adhesive probe was placed on the right index finger, which immediately registered a reading of 99%–100% with the patient still receiving an Fio2 of 29% (Fig. 3).
The pulse oximeter remained connected to the new probe on the right index finger with a reading of 99% while using an Fio2 of 29%. The biomedical engineering group was summoned to investigate this significant discrepancy. When the engineers arrived, the pulse oximeter cable was reattached to the first probe on the left middle finger. The Spo2 immediately changed to 94%, but the signal-to-noise ratio indicator remained at 3 stars. Finally, a third pulse oximeter probe was applied to the left fifth digit, which gave a saturation reading of 99% with a 3-star signal quality indicator (Fig. 4). Thus, the left fifth digit Spo2 registered 99%, but the original pulse oximeter probe 2 fingers over, on the middle finger of the same hand, registered 94% seconds earlier. It was therefore decided to take the first probe from the left middle finger to the engineering workshop for further examination. Before removing the probe, it was noticed that the first pulse oximeter probe (on the left middle finger) had been misapplied, with a misalignment distance of >20 mm (Fig. 5, left side). This same probe was then removed and reapplied to the patient’s finger in a normal fashion whereupon it promptly gave an Spo2 of 99%.
Misalignment Distance Measurements in Postanesthesia Care Unit Patients
In our 3 cases, the “misalignment distance” was 15, 10, and 23 mm, respectively. To estimate the incidence and the severity of probe misalignment in our practice, we measured the probe misalignment distance from a spot sample of 100 postanesthesia care unit (PACU) patients. The disposable pulse oximeter probes were placed either by anesthesia residents or nurses. We found that the average misalignment distance was 5.4 mm, with a standard deviation of 4.7 mm and a range of 0–23 mm. Only 6% of patients had perfect alignment with a misalignment distance of 0 mm. Of the remaining 94 patients, 47% had misalignment distances of 1–4 mm, 26% had misalignment distances of 5–9 mm, and 21% had misalignment distances of 10 mm or more. None of these PACU patients had Spo2 readings low enough to require treatment other than routine nasal cannulae or facemask oxygen supplementation.
Malposition of reusable pulse oximeter probes has been reported to yield erroneously low saturation values, and the low readings are attributed to a phenomenon known as the penumbra effect.3–5 This happens, for example, when a reusable probe partially falls off the finger, so that the light is shunted from the emitting LED directly to the photodetector. This will cause a falsely low Spo2 if the actual SaO2 is >85% and a falsely increased Spo2 if the actual SaO2 is <85%. Importantly, the quality of the plethysmographic tracing usually correlates with probe malposition. Herein, we have described 3 cases in which misaligned disposable pulse oximeter probes reported falsely low saturations despite excellent-quality (3-star) plethysmographic tracings in each case.
The low saturation in combination with a high-quality plethysmograph led to unnecessary intraoperative bronchoscopies (all cases), changes of intraoperative anesthesia management (Cases 1 and 2), and unnecessary biomedical engineering involvement (Case 3). In all 3 cases, the problem was initially masked by good plethysmographic tracings, high signal quality indices, and reasonable saturation readings during oxygen administration at induction. The high Spo2 and good signal quality indications during administration of oxygen in all 3 cases gave anesthesiologists the false sense of security that if the disposable probe provided a normal saturation reading during oxygen administration, then it would continue to work well during the case because unlike the reusable probes that could easily become dislodged, the adhesive disposable probes would not change positions once they were put on. As a result, when low saturation readings occurred during the case, anesthesiologists tended to focus on other causes for the apparent hypoxia. Only after other possible causes of hypoxia were excluded did they check the pulse oximeter probe itself as a cause of apparent hypoxia. Retrospectively, an early investigation to ascertain the proper positioning of the probe would have resulted in fewer unnecessary interventions. In our cases, the interventions were not apparently harmful. Nevertheless, it still seems prudent to start the investigation of low Spo2 by making a quick check for sensor misalignment, while simultaneously moving to exclude actual cases of hypoxia.
There might be several reasons why the disposable probes were not applied perfectly in our practice. First, malpositioned disposable probes often yield a high-quality plethysmographic tracing, and therefore many cases of misaligned probes are not detected because the data appear believable. In our survey of 100 PACU patients, only 1 of the 21 patients with misalignment of 10 mm or more had intraoperative saturation readings that were low enough to capture the anesthesia team’s attention. Second, factors such as long fingernails, as in Figure 1, often make proper placement of disposable probes difficult. Third, if the probe is misapplied, the adhesive makes it difficult to take the probe off and adjust it once it is applied to the patient’s finger. Fourth, the probe appears to be a simple and self-explanatory device that needs no particular care during application. Intuitively, such misalignment potential is a general feature of all disposable pulse oximeter probes.
Ideal placement of disposable pulse oximeter probes occurs when the photodetector is applied directly opposite to the LED. However, there is no report in the literature demonstrating how precisely or “correctly” disposable probes are actually applied. In our practice, only 6% of disposable probes were applied with perfect alignment of the LED and photodetector with a misalignment distance of 0 mm, whereas a little more than half of all probes were applied with a misalignment distance of <0.5 cm. In our setting, 1 in every 5 disposable probes is applied with the photodetector displaced by 10 mm or more from ideal alignment.
Apparently, misaligned disposable pulse oximeter sensors do not always give a false saturation reading. In our survey of 100 PACU patients, none of the sensors with misalignment distances of <10 mm and fewer than 5% of sensors with misalignment distances of 10 mm or more required intraoperative attention or sensor replacement. Our results show that such perfect alignment is infrequently achieved in practice. However, there is no guideline regarding how much misalignment still provides good performance when a disposable pulse oximeter probe is applied. Further study is therefore needed to determine this safety zone, and to determine the factors that can cause misreading of misaligned probes. In addition, basic education is required to teach practitioners, including anesthesia residents, certified registered nurse anesthetists, and operating room nurses how to correctly apply disposable pulse oximeter probes and how to recognize the misaligned probes.
Our report indicates that misplaced disposable oximeter probes can provide high-quality plethysmographic signals and yield believable, yet incorrect, saturation readings. As a result, when a disposable probe is used, the first step in the investigation of apparent hypoxia should be to search for real causes, but also to simultaneously check to see if the probe is properly applied. Moreover, our report further demonstrates that although the signal quality indication may be necessary to document probe performance, it is not sufficient to assure the performance of the oximeter.
2.Dorsch JA, Dorsch SE. Understanding anesthesia equipment. 5th ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2007
3.Kelleher JF, Ruff RH. The penumbra effect: vasomotion-dependent pulse oximeter artifact due to probe malposition. Anesthesiology 1989;71:787–91
4.Southall DP, Samuels M. Inappropriate sensor application in pulse oximetry. Lancet 1992;340:481–2
5.Barker SJ, Hyatt J, Shah NK, Kao YJ. The effect of sensor malpositioning on pulse oximeter accuracy during hypoxemia. Anesthesiology 1993;79:248–54